diff --git a/CHANGELOG.md b/CHANGELOG.md index 28bcaae5b6..6fce19483c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -20,6 +20,7 @@ Documentation for Composable Kernel available at [https://rocm.docs.amd.com/proj * Added tensor-wise quantization for CK_TILE GEMM. * Added support for batched contraction kernel. * Added pooling kernel in CK_TILE +* Added top-k sigmoid kernel in CK_TILE ### Changed diff --git a/Jenkinsfile b/Jenkinsfile index c642e2d3b1..625b6d0d09 100644 --- a/Jenkinsfile +++ b/Jenkinsfile @@ -194,6 +194,33 @@ def check_arch(){ return arch_type } +def check_arch_name(){ + def arch_name = "" + sh 'rocminfo | tee rocminfo.log' + if ( runShell('grep -n "gfx90a" rocminfo.log') ){ + arch_name = "gfx90a" + } + else if ( runShell('grep -n "gfx942" rocminfo.log') ) { + arch_name = "gfx942" + } + else if ( runShell('grep -n "gfx10" rocminfo.log') ) { + arch_name = "gfx10" + } + else if ( runShell('grep -n "gfx11" rocminfo.log') ) { + arch_name = "gfx11" + } + else if ( runShell('grep -n "gfx12" rocminfo.log') ) { + arch_name = "gfx12" + } + else if ( runShell('grep -n "gfx908" rocminfo.log') ) { + arch_name = "gfx908" + } + else if ( runShell('grep -n "gfx950" rocminfo.log') ) { + arch_name = "gfx950" + } + return arch_name +} + def getDockerImage(Map conf=[:]){ env.DOCKER_BUILDKIT=1 def prefixpath = conf.get("prefixpath", "/opt/rocm") @@ -302,12 +329,6 @@ def cmake_build(Map conf=[:]){ //cmake_env can overwrite default CXX variables. def cmake_envs = "CXX=${compiler} CXXFLAGS='-Werror' " + conf.get("cmake_ex_env","") - def package_build = (conf.get("package_build","") == "true") - - if (package_build == true) { - config_targets = "package" - } - if(conf.get("build_install","") == "true") { config_targets = 'install ' + config_targets @@ -455,15 +476,20 @@ def cmake_build(Map conf=[:]){ else{ sh "ninja check" } + if(params.BUILD_PACKAGES){ + echo "Build ckProfiler packages" + sh 'ninja -j64 package' + def arch_name = check_arch_name() + sh "mv composablekernel-ckprofiler_*.deb composablekernel-ckprofiler_1.2.0_amd64_${arch_name}.deb" + stash includes: "composablekernel-ckprofiler**.deb", name: "profiler_package_${arch_name}" + } } if(params.BUILD_INSTANCES_ONLY){ // build deb packages - echo "Build packages" + echo "Build library package" sh 'ninja -j64 package' - archiveArtifacts artifacts: 'composablekernel-dev*.deb' sh 'mv composablekernel-dev_*.deb composablekernel-dev_all_targets_1.2.0_amd64.deb' - sh 'mv composablekernel-ckprofiler_*.deb composablekernel-ckprofiler_1.2.0_amd64.deb' - stash includes: "composablekernel-**.deb", name: "packages" + stash includes: "composablekernel-dev**.deb", name: "lib_package" } } else{ @@ -475,15 +501,18 @@ def cmake_build(Map conf=[:]){ else{ sh "ninja check" } + if(params.BUILD_PACKAGES){ + echo "Build ckProfiler packages" + sh 'ninja -j64 package' + def arch_name = check_arch_name() + sh "mv composablekernel-ckprofiler_*.deb composablekernel-ckprofiler_1.2.0_amd64_${arch_name}.deb" + stash includes: "composablekernel-ckprofiler**.deb", name: "profiler_package_${arch_name}" + } } } } } - // Only archive from develop - if (package_build == true && env.BRANCH_NAME == "develop") { - archiveArtifacts artifacts: "build/*.deb", allowEmptyArchive: true, fingerprint: true - } //check the node gpu architecture def arch = check_arch() if (params.RUN_CK_TILE_FMHA_TESTS){ @@ -823,9 +852,42 @@ def process_results(Map conf=[:]){ } if (params.BUILD_INSTANCES_ONLY){ // unstash deb packages - unstash "packages" + try{ + unstash "lib_package" + } + catch(Exception err){ + echo "could not locate lib_package." + } sh "sshpass -p ${env.ck_deb_pw} scp -o StrictHostKeyChecking=no composablekernel-*.deb ${env.ck_deb_user}@${env.ck_deb_ip}:/var/www/html/composable_kernel/" } + if (params.BUILD_PACKAGES){ + // unstash deb packages + try{ + unstash "profiler_package_gfx90a" + } + catch(Exception err){ + echo "could not locate profiler_package_gfx90a." + } + try{ + unstash "profiler_package_gfx942" + } + catch(Exception err){ + echo "could not locate profiler_package_gfx942." + } + try{ + unstash "profiler_package_gfx950" + } + catch(Exception err){ + echo "could not locate profiler_package_gfx950." + } + try{ + unstash "profiler_package_gfx12" + } + catch(Exception err){ + echo "could not locate profiler_package_gfx12." + } + sh "sshpass -p ${env.ck_deb_pw} scp -o StrictHostKeyChecking=no composablekernel-ckprofiler*.deb ${env.ck_deb_user}@${env.ck_deb_ip}:/var/www/html/composable_kernel/" + } else{ // unstash perf files to master try{ @@ -993,7 +1055,7 @@ def run_pytorch_tests(Map conf=[:]){ //launch develop branch daily jobs CRON_SETTINGS = BRANCH_NAME == "develop" ? '''0 23 * * * % RUN_FULL_QA=true;RUN_CK_TILE_FMHA_TESTS=true;RUN_PERFORMANCE_TESTS=true;FORCE_CI=true 0 22 * * * % RUN_FULL_QA=true;DISABLE_DL_KERNELS=true;RUN_TILE_ENGINE_GEMM_TESTS=true;RUN_PERFORMANCE_TESTS=true;RUN_ALL_UNIT_TESTS=true;FORCE_CI=true - 0 21 * * * % RUN_GROUPED_CONV_LARGE_CASES_TESTS=true;hipTensor_test=true;BUILD_GFX908=true;BUILD_GFX942=true;BUILD_GFX950=true;RUN_PERFORMANCE_TESTS=true;RUN_ALL_UNIT_TESTS=true;FORCE_CI=true + 0 21 * * * % RUN_GROUPED_CONV_LARGE_CASES_TESTS=true;hipTensor_test=true;BUILD_GFX908=true;BUILD_GFX942=true;BUILD_GFX950=true;RUN_PERFORMANCE_TESTS=true;RUN_ALL_UNIT_TESTS=true;FORCE_CI=true;BUILD_PACKAGES=true 0 19 * * * % BUILD_DOCKER=true;COMPILER_VERSION=amd-staging;BUILD_COMPILER=/llvm-project/build/bin/clang++;USE_SCCACHE=false;NINJA_BUILD_TRACE=true;RUN_ALL_UNIT_TESTS=true;FORCE_CI=true 0 17 * * * % BUILD_DOCKER=true;COMPILER_VERSION=amd-mainline;BUILD_COMPILER=/llvm-project/build/bin/clang++;USE_SCCACHE=false;NINJA_BUILD_TRACE=true;RUN_ALL_UNIT_TESTS=true;FORCE_CI=true 0 15 * * * % BUILD_INSTANCES_ONLY=true;USE_SCCACHE=false;NINJA_BUILD_TRACE=true;FORCE_CI=true @@ -1085,6 +1147,10 @@ pipeline { name: "BUILD_INSTANCES_ONLY", defaultValue: false, description: "Test building instances for various architectures simultaneously (default: OFF)") + booleanParam( + name: "BUILD_PACKAGES", + defaultValue: false, + description: "Build packages for the libraries and/or ckProfiler (default: OFF)") booleanParam( name: "BUILD_GFX908", defaultValue: false, @@ -1829,7 +1895,7 @@ pipeline { stage("Process results"){ when { beforeAgent true - expression { (params.RUN_PERFORMANCE_TESTS.toBoolean() || params.BUILD_INSTANCES_ONLY.toBoolean() || params.RUN_CK_TILE_FMHA_TESTS.toBoolean()) && !params.BUILD_LEGACY_OS.toBoolean() } + expression { (params.RUN_PERFORMANCE_TESTS.toBoolean() || params.BUILD_INSTANCES_ONLY.toBoolean() || params.RUN_CK_TILE_FMHA_TESTS.toBoolean()|| params.BUILD_PACKAGES.toBoolean()) && !params.BUILD_LEGACY_OS.toBoolean() } } agent { label 'mici' } steps{ diff --git a/example/15_grouped_gemm/grouped_gemm_xdl_fp32.cpp b/example/15_grouped_gemm/grouped_gemm_xdl_fp32.cpp index fb047ae364..236e5e4fa2 100644 --- a/example/15_grouped_gemm/grouped_gemm_xdl_fp32.cpp +++ b/example/15_grouped_gemm/grouped_gemm_xdl_fp32.cpp @@ -59,4 +59,11 @@ using DeviceGemmInstance = ck::tensor_operation::device::DeviceGroupedGemm_Xdl #include "run_grouped_gemm_example.inc" -int main(int argc, char* argv[]) { return !run_grouped_gemm_example(argc, argv); } +int main(int argc, char* argv[]) +{ + if(ck::is_gfx11_supported() || ck::is_gfx12_supported()) + { + return 0; + } + return !run_grouped_gemm_example(argc, argv); +} diff --git a/example/15_grouped_gemm/run_grouped_gemm_example.inc b/example/15_grouped_gemm/run_grouped_gemm_example.inc index 4ef6074f4a..87ccebc3c4 100644 --- a/example/15_grouped_gemm/run_grouped_gemm_example.inc +++ b/example/15_grouped_gemm/run_grouped_gemm_example.inc @@ -278,19 +278,20 @@ bool run_grouped_gemm_example(int argc, char* argv[]) problem_size.group_count = 16; - if(argc == 4) + if(argc == 1) + { + // use default cases + } + else if(argc == 4 || argc == 6) { config.do_verification = std::stoi(argv[1]); config.init_method = std::stoi(argv[2]); config.time_kernel = std::stoi(argv[3]); - } - else if(argc == 6) - { - config.do_verification = std::stoi(argv[1]); - config.init_method = std::stoi(argv[2]); - config.time_kernel = std::stoi(argv[3]); - config.async_hargs = std::stoi(argv[4]); - problem_size.group_count = std::stoi(argv[5]); + if(argc == 6) + { + config.async_hargs = std::stoi(argv[4]); + problem_size.group_count = std::stoi(argv[5]); + } } else { @@ -299,18 +300,33 @@ bool run_grouped_gemm_example(int argc, char* argv[]) printf("arg3: time kernel (0=n0, 1=yes)\n"); printf("arg4: async hargs (0=n0, 1=yes)\n"); printf("arg5: group count (default=16)"); - exit(0); + exit(1); } + // Lambda to get stride based on layout + auto get_stride = [](auto layout, auto row_dim, auto col_dim) { + if constexpr(std::is_same_v) + { + return col_dim; + } + else + { + return row_dim; + } + }; + for(int i = 0; i < problem_size.group_count; i++) { problem_size.Ms.push_back(256 + 256 * i); problem_size.Ns.push_back(128 + 128 * i); problem_size.Ks.push_back(128 + 64 * i); - problem_size.stride_As.push_back(problem_size.Ks[i]); - problem_size.stride_Bs.push_back(problem_size.Ks[i]); - problem_size.stride_Cs.push_back(problem_size.Ns[i]); + problem_size.stride_As.push_back( + get_stride(ALayout{}, problem_size.Ms[i], problem_size.Ks[i])); + problem_size.stride_Bs.push_back( + get_stride(BLayout{}, problem_size.Ks[i], problem_size.Ns[i])); + problem_size.stride_Cs.push_back( + get_stride(ELayout{}, problem_size.Ms[i], problem_size.Ns[i])); } return run_grouped_gemm(problem_size, config); diff --git a/example/45_elementwise_normalization/elementwise_layernorm_blockwise.cpp b/example/45_elementwise_normalization/elementwise_layernorm_blockwise.cpp index 8064809123..21c5ff8d5a 100644 --- a/example/45_elementwise_normalization/elementwise_layernorm_blockwise.cpp +++ b/example/45_elementwise_normalization/elementwise_layernorm_blockwise.cpp @@ -82,37 +82,29 @@ int main(int argc, char* argv[]) bool do_verification = true; bool time_kernel = true; + ck::index_t M = 48 * 256; + ck::index_t N = 1024; + if(argc == 1) { // use default } - else if(argc == 3) + else if(argc == 3 || argc == 5) { do_verification = std::stoi(argv[1]); time_kernel = std::stoi(argv[2]); + if(argc == 5) + { + M = std::stoi(argv[3]); + N = std::stoi(argv[4]); + } } else { printf("arg1: verification (0=no, 1=yes)\n"); printf("arg2: time kernel (0=no, 1=yes)\n"); - exit(0); - } - - ck::index_t M = 48 * 256; - ck::index_t N = 1024; - if(argc == 1) - { - // use default case - } - else if(argc == 3) - { - M = std::stoi(argv[1]); - N = std::stoi(argv[2]); - } - else - { - std::cerr << "arg1 to 2: M, N" << std::endl; - return 1; + printf("arg3-4: M, N\n"); + exit(1); } ck::index_t Stride = N; diff --git a/example/ck_tile/09_topk_softmax/topk_softmax.cpp b/example/ck_tile/09_topk_softmax/topk_softmax.cpp index 0487bd05d2..400329986a 100644 --- a/example/ck_tile/09_topk_softmax/topk_softmax.cpp +++ b/example/ck_tile/09_topk_softmax/topk_softmax.cpp @@ -83,6 +83,26 @@ auto reference_topk_softmax(const ck_tile::HostTensor& x, reference_topk(y, y_values, y_indices, k, dim, largest, sorted); } +template +auto reference_topk_sigmoid(const ck_tile::HostTensor& x, + ck_tile::HostTensor& y_values, + ck_tile::HostTensor& y_indices, + ck_tile::index_t k, + ck_tile::index_t dim = -1, + bool largest = true, + bool sorted = true) +{ + using namespace ck_tile; + + // topk only - no need to apply the sigmoid first + auto x_fp32 = x.template CopyAsType(); + reference_topk(x_fp32, y_values, y_indices, k, dim, largest, sorted); + // apply sigmoid + std::transform(y_values.begin(), y_values.end(), y_values.begin(), [](auto value) { + return WeightType(1) / (WeightType(1) + exp(-value)); + }); +} + // different threshold for different dtype template auto get_elimit(std::string /*init_method*/) @@ -133,7 +153,8 @@ auto create_args(int argc, char* argv[]) .insert("warmup", "5", "number of iterations before benchmark the kernel") .insert("repeat", "20", "number of iterations to benchmark the kernel") .insert("json", "0", "0: No Json, 1: Dump Results in Json format") - .insert("jsonfile", "topk_softmax.json", "json file name to dump results"); + .insert("jsonfile", "topk_softmax.json", "json file name to dump results") + .insert("activation", "softmax", "activation function to use: softmax or sigmoid"); bool result = arg_parser.parse(argc, argv); return std::make_tuple(result, arg_parser); @@ -154,6 +175,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) int kname = args.get_int("kname"); int warmup = args.get_int("warmup"); int repeat = args.get_int("repeat"); + std::string activation = args.get_str("activation"); if(stride_input < 0) { @@ -204,7 +226,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) x_dev.ToDevice(x_host.data()); - topk_softmax_trait trait{input_prec, weight_prec, experts}; + topk_softmax_trait trait{input_prec, weight_prec, experts, activation}; topk_softmax_kargs karg{x_dev.GetDeviceBuffer(), value_dev.GetDeviceBuffer(), @@ -221,7 +243,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) warmup, repeat}; auto ms = topk_softmax(trait, karg, sc); - printf("[%s|%s]tokens:%d, experts:%d, topk:%d, st_i:%d, st_o:%d, ms:%f, ", + printf("[%s|%s]tokens:%d, experts:%d, topk:%d, st_i:%d, st_o:%d, activation:%s, ms:%f, ", input_prec.c_str(), weight_prec.c_str(), tokens, @@ -229,6 +251,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) topk, stride_input, stride_output, + activation.c_str(), ms); if(ms < 0) printf("not supported\n"); @@ -247,8 +270,20 @@ bool test_topk_softmax(ck_tile::ArgParser args) ck_tile::HostTensor value_ref({tokens, topk}, {stride_output, 1}); ck_tile::HostTensor index_ref({tokens, topk}, {stride_output, 1}); - reference_topk_softmax( - x_host, value_ref, index_ref, topk); + if(activation == "softmax") + { + reference_topk_softmax( + x_host, value_ref, index_ref, topk); + } + else if(activation == "sigmoid") + { + reference_topk_sigmoid( + x_host, value_ref, index_ref, topk); + } + else + { + throw std::runtime_error("unsupported activation type: " + activation); + } auto [rtol, atol] = get_elimit(""); for(int i_t = 0; i_t < tokens; i_t++) diff --git a/example/ck_tile/09_topk_softmax/topk_softmax_api.cpp b/example/ck_tile/09_topk_softmax/topk_softmax_api.cpp index 6e6bb20020..bdcfb47cc7 100644 --- a/example/ck_tile/09_topk_softmax/topk_softmax_api.cpp +++ b/example/ck_tile/09_topk_softmax/topk_softmax_api.cpp @@ -3,27 +3,31 @@ #include "topk_softmax_api.hpp" -#define TOPK_SOFTMAX_DISPATCH(experts_) \ - constexpr ck_tile::index_t ts_experts = experts_; \ - using ts_problem = ck_tile:: \ - TopkSoftmaxWarpPerRowProblem; \ - using ts_pipeline = ck_tile::TopkSoftmaxWarpPerRowPipeline; \ - \ - using kernel = ck_tile::TopkSoftmaxKernel; \ - \ - auto kargs = kernel::MakeKargs(a); \ - \ - const dim3 grids = kernel::GridSize(a); \ - const dim3 blocks = kernel::BlockSize(); \ - \ - float ave_time = \ - ck_tile::launch_kernel(s, ck_tile::make_kernel<1>(kernel{}, grids, blocks, 0, kargs)); \ - \ +#define TOPK_SOFTMAX_DISPATCH(experts_, use_softmax_) \ + constexpr ck_tile::index_t ts_experts = experts_; \ + constexpr bool ts_use_softmax = use_softmax_; \ + using ts_problem = ck_tile::TopkSoftmaxWarpPerRowProblem; \ + using ts_pipeline = ck_tile::TopkSoftmaxWarpPerRowPipeline; \ + \ + using kernel = ck_tile::TopkSoftmaxKernel; \ + \ + auto kargs = kernel::MakeKargs(a); \ + \ + const dim3 grids = kernel::GridSize(a); \ + const dim3 blocks = kernel::BlockSize(); \ + \ + float ave_time = \ + ck_tile::launch_kernel(s, ck_tile::make_kernel<1>(kernel{}, grids, blocks, 0, kargs)); \ + \ return ave_time; float topk_softmax(topk_softmax_trait t, topk_softmax_kargs a, ck_tile::stream_config s) { - if(t.input_type == "fp16" && t.weight_type == "fp32") + if(t.input_type == "fp16" && t.weight_type == "fp32" && t.activation == "softmax") { using ts_input_type = ck_tile::fp16_t; using ts_weight_type = float; @@ -31,36 +35,36 @@ float topk_softmax(topk_softmax_trait t, topk_softmax_kargs a, ck_tile::stream_c #if 1 if(t.experts <= 8) { - TOPK_SOFTMAX_DISPATCH(8) + TOPK_SOFTMAX_DISPATCH(8, true) } else if(t.experts <= 16) { - TOPK_SOFTMAX_DISPATCH(16) + TOPK_SOFTMAX_DISPATCH(16, true) } else if(t.experts <= 32) { - TOPK_SOFTMAX_DISPATCH(32) + TOPK_SOFTMAX_DISPATCH(32, true) } else if(t.experts <= 64) { - TOPK_SOFTMAX_DISPATCH(64) + TOPK_SOFTMAX_DISPATCH(64, true) } else if(t.experts <= 128) { - TOPK_SOFTMAX_DISPATCH(128) + TOPK_SOFTMAX_DISPATCH(128, true) } else if(t.experts <= 192) { - TOPK_SOFTMAX_DISPATCH(192) + TOPK_SOFTMAX_DISPATCH(192, true) } #else if(t.experts <= 128) { - TOPK_SOFTMAX_DISPATCH(128) + TOPK_SOFTMAX_DISPATCH(128, true) } #endif } - else if(t.input_type == "bf16" && t.weight_type == "fp32") + else if(t.input_type == "bf16" && t.weight_type == "fp32" && t.activation == "softmax") { #if 1 using ts_input_type = ck_tile::bf16_t; @@ -68,27 +72,96 @@ float topk_softmax(topk_softmax_trait t, topk_softmax_kargs a, ck_tile::stream_c using ts_index_type = ck_tile::index_t; if(t.experts <= 8) { - TOPK_SOFTMAX_DISPATCH(8) + TOPK_SOFTMAX_DISPATCH(8, true) } else if(t.experts <= 16) { - TOPK_SOFTMAX_DISPATCH(16) + TOPK_SOFTMAX_DISPATCH(16, true) } else if(t.experts <= 32) { - TOPK_SOFTMAX_DISPATCH(32) + TOPK_SOFTMAX_DISPATCH(32, true) } else if(t.experts <= 64) { - TOPK_SOFTMAX_DISPATCH(64) + TOPK_SOFTMAX_DISPATCH(64, true) } else if(t.experts <= 128) { - TOPK_SOFTMAX_DISPATCH(128) + TOPK_SOFTMAX_DISPATCH(128, true) } else if(t.experts <= 192) { - TOPK_SOFTMAX_DISPATCH(192) + TOPK_SOFTMAX_DISPATCH(192, true) + } +#endif + } + else if(t.input_type == "fp16" && t.weight_type == "fp32" && t.activation == "sigmoid") + { + using ts_input_type = ck_tile::fp16_t; + using ts_weight_type = float; + using ts_index_type = ck_tile::index_t; +#if 1 + if(t.experts <= 8) + { + TOPK_SOFTMAX_DISPATCH(8, false) + } + else if(t.experts <= 16) + { + TOPK_SOFTMAX_DISPATCH(16, false) + } + else if(t.experts <= 32) + { + TOPK_SOFTMAX_DISPATCH(32, false) + } + else if(t.experts <= 64) + { + TOPK_SOFTMAX_DISPATCH(64, false) + } + else if(t.experts <= 128) + { + TOPK_SOFTMAX_DISPATCH(128, false) + } + else if(t.experts <= 192) + { + TOPK_SOFTMAX_DISPATCH(192, false) + } +#else + if(t.experts <= 128) + { + TOPK_SOFTMAX_DISPATCH(128, false) + } +#endif + } + else if(t.input_type == "bf16" && t.weight_type == "fp32" && t.activation == "sigmoid") + { +#if 1 + using ts_input_type = ck_tile::bf16_t; + using ts_weight_type = float; + using ts_index_type = ck_tile::index_t; + if(t.experts <= 8) + { + TOPK_SOFTMAX_DISPATCH(8, false) + } + else if(t.experts <= 16) + { + TOPK_SOFTMAX_DISPATCH(16, false) + } + else if(t.experts <= 32) + { + TOPK_SOFTMAX_DISPATCH(32, false) + } + else if(t.experts <= 64) + { + TOPK_SOFTMAX_DISPATCH(64, false) + } + else if(t.experts <= 128) + { + TOPK_SOFTMAX_DISPATCH(128, false) + } + else if(t.experts <= 192) + { + TOPK_SOFTMAX_DISPATCH(192, false) } #endif } diff --git a/example/ck_tile/09_topk_softmax/topk_softmax_api.hpp b/example/ck_tile/09_topk_softmax/topk_softmax_api.hpp index 65651efa4d..c98a887736 100644 --- a/example/ck_tile/09_topk_softmax/topk_softmax_api.hpp +++ b/example/ck_tile/09_topk_softmax/topk_softmax_api.hpp @@ -12,6 +12,7 @@ struct topk_softmax_trait std::string input_type; std::string weight_type; // currently always float int experts; + std::string activation; // "softmax" or "sigmoid" }; struct topk_softmax_kargs : public ck_tile::TopkSoftmaxHostArgs diff --git a/experimental/builder/test/CMakeLists.txt b/experimental/builder/test/CMakeLists.txt index 1dc508a0c3..bafa95862a 100644 --- a/experimental/builder/test/CMakeLists.txt +++ b/experimental/builder/test/CMakeLists.txt @@ -2,7 +2,7 @@ include(gtest) # Helper function to create a gtest executable with common properties function(add_ck_builder_test test_name) - add_executable(${test_name} ${ARGN}) + add_executable(${test_name} ${ARGN} testing_utils.cpp) target_compile_features(${test_name} PRIVATE cxx_std_20) target_include_directories(${test_name} PRIVATE "${PROJECT_SOURCE_DIR}/experimental/builder/include" @@ -20,9 +20,28 @@ endfunction() add_ck_builder_test(test_conv_builder test_conv_builder.cpp test_instance_traits.cpp - test_instance_traits_util.cpp - testing_utils.cpp) + test_instance_traits_util.cpp) # Testing the virtual GetInstanceString methods requires kernel compilation. add_ck_builder_test(test_get_instance_string test_get_instance_string.cpp) + +add_ck_builder_test(test_inline_diff test_inline_diff.cpp) + +function(add_ck_factory_test test_name) + add_ck_builder_test(${test_name} ${ARGN}) + target_link_libraries(${test_name} PRIVATE composablekernels::device_conv_operations) +endfunction() + +add_ck_factory_test(test_testing_utils test_testing_utils.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward test_ck_factory_grouped_convolution_forward.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_clamp test_ck_factory_grouped_convolution_forward_clamp.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_convscale test_ck_factory_grouped_convolution_forward_convscale.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_bilinear test_ck_factory_grouped_convolution_forward_bilinear.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_scale test_ck_factory_grouped_convolution_forward_scale.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_scaleadd_ab test_ck_factory_grouped_convolution_forward_scaleadd_ab.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_bias_clamp test_ck_factory_grouped_convolution_forward_bias_clamp.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_bias_bnorm_clamp test_ck_factory_grouped_convolution_forward_bias_bnorm_clamp.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_scaleadd_scaleadd_relu test_ck_factory_grouped_convolution_forward_scaleadd_scaleadd_relu.cpp) +add_ck_factory_test(test_ck_factory_grouped_convolution_forward_dynamic_op test_ck_factory_grouped_convolution_forward_dynamic_op.cpp) + diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward.cpp new file mode 100644 index 0000000000..53199f1d77 --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward.cpp @@ -0,0 +1,2345 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +using namespace ck::tensor_layout::convolution; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::PassThrough; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD, + OutLayout, + InDataType, + WeiDataType, + ck::Tuple<>, + OutDataType, + PassThrough, + PassThrough, + PassThrough, + AComputeType, + BComputeType>; + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +// 1D kernels - layout GNWC/GKXC/GNWK + +struct F32_1D_GNWC_GKXC_GNWK +{ + using DeviceOp = ::DeviceOp<1, GNWC, GKXC, GNWK, float>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16_1D_GNWC_GKXC_GNWK +{ + using DeviceOp = ::DeviceOp<1, GNWC, GKXC, GNWK, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16_1D_GNWC_GKXC_GNWK +{ + using DeviceOp = ::DeviceOp<1, GNWC, GKXC, GNWK, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8_1D_GNWC_GKXC_GNWK +{ + using DeviceOp = ::DeviceOp<1, GNWC, GKXC, GNWK, int8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +// 2D kernels - layout GNHWC/GKYXC/GNHWK + +struct F32_2D_GNHWC_GKYXC_GNHWK +{ + using DeviceOp = ::DeviceOp<2, GNHWC, GKYXC, GNHWK, float>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16_2D_GNHWC_GKYXC_GNHWK +{ + using DeviceOp = ::DeviceOp<2, GNHWC, GKYXC, GNHWK, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16_2D_GNHWC_GKYXC_GNHWK +{ + using DeviceOp = ::DeviceOp<2, GNHWC, GKYXC, GNHWK, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8_2D_GNHWC_GKYXC_GNHWK +{ + using DeviceOp = ::DeviceOp<2, GNHWC, GKYXC, GNHWK, int8_t>; + + constexpr static std::initializer_list expected = { + // clang-format off + // clang-format on + }; +}; + +// 2D kernels - layout NHWGC/GKYXC/NHWGK + +struct F32_2D_NHWGC_GKYXC_NHWGK +{ + using DeviceOp = ::DeviceOp<2, NHWGC, GKYXC, NHWGK, float>; + + constexpr static auto expected = { + // clang-format off + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "" + // clang-format on + }; +}; + +struct F16_2D_NHWGC_GKYXC_NHWGK +{ + using DeviceOp = ::DeviceOp<2, NHWGC, GKYXC, NHWGK, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "" + // clang-format on + }; +}; + +struct BF16_2D_NHWGC_GKYXC_NHWGK +{ + using DeviceOp = ::DeviceOp<2, NHWGC, GKYXC, NHWGK, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "" + // clang-format on + }; +}; + +struct S8_2D_NHWGC_GKYXC_NHWGK +{ + using DeviceOp = ::DeviceOp<2, NHWGC, GKYXC, NHWGK, int8_t>; + + constexpr static auto expected = { + // clang-format off + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,s8,s8>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,s8,s8,s32,s8,EmptyTuple,s8,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,s8,s8>", + "" + // clang-format on + }; +}; + +// 2D kernels - layout NGCHW/GKCYX/NGKHW + +struct F32_2D_NGCHW_GKCYX_NGKHW +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKCYX, NGKHW, float>; + + constexpr static auto expected = { + // clang-format off + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "" + // clang-format on + }; +}; + +struct F16_2D_NGCHW_GKCYX_NGKHW +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKCYX, NGKHW, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "" + // clang-format on + }; +}; + +struct BF16_2DNGCHW_GKCYX_NGKHWK +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKCYX, NGKHW, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NGCHW,GKCYX,EmptyTuple,NGKHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "" + // clang-format on + }; +}; + +// 2D kernels - layout NGCHW/GKYXC/NGKHW + +struct F32_2D_NGCHW_GKYXC_NGKHW +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKYXC, NGKHW, float>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16_2D_NGCHW_GKYXC_NGKHW +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKYXC, NGKHW, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16_2D_NGCHW_GKYXC_NGKHW +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKYXC, NGKHW, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8_2D_NGCHW_GKYXC_NGKHW +{ + using DeviceOp = ::DeviceOp<2, NGCHW, GKYXC, NGKHW, int8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +// 3D - layout GNDHWC/GKZYXC/GNDHWK + +struct F32_3D_GNDHWC_GKZYXC_GNDHWK +{ + using DeviceOp = ::DeviceOp<3, GNDHWC, GKZYXC, GNDHWK, float>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16_3D_GNDHWC_GKZYXC_GNDHWK +{ + using DeviceOp = ::DeviceOp<3, GNDHWC, GKZYXC, GNDHWK, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16_3D_GNDHWC_GKZYXC_GNDHWK +{ + using DeviceOp = ::DeviceOp<3, GNDHWC, GKZYXC, GNDHWK, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8_3D_GNDHWC_GKZYXC_GNDHWK +{ + using DeviceOp = ::DeviceOp<3, GNDHWC, GKZYXC, GNDHWK, int8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +// 3D - layout NDHWGC/GKZYXC/NDHWGK + +struct F32_3D_NDHWGC_GKZYXC_NDHWGK_TF32 +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, float, float, float, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F32_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, float>; + + constexpr static auto expected = { + // clang-format off + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "" + // clang-format on + }; +}; + +struct F16_3D_NDHWGC_GKZYXC_NDHWGK_F8 +{ + using DeviceOp = + ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::half_t, ck::half_t, ck::half_t, ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF8_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::bf8_t, ck::bf8_t, ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_BF8_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = + ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::f8_t, ck::bf8_t, ck::f8_t, ck::f8_t, ck::bf8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF8_F8_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = + ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::bf8_t, ck::f8_t, ck::f8_t, ck::bf8_t, ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +struct S8_3D_NDHWGC_GKZYXC_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NDHWGC, GKZYXC, NDHWGK, int8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +// 3D - layout NGCDHW/GKCZYX/NGKDHW + +struct F32_3D_NGCDHW_GKCZYX_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NGCDHW, GKCZYX, NGKDHW, float>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F16_3D_NGCDHW_GKCZYX_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NGCDHW, GKCZYX, NGKDHW, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_3D_NGCDHW_GKCZYX_NDHWGK +{ + using DeviceOp = ::DeviceOp<3, NGCDHW, GKCZYX, NGKDHW, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NGCDHW,GKCZYX,EmptyTuple,NGKDHW,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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +using TestTypes = ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bias_bnorm_clamp.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bias_bnorm_clamp.cpp new file mode 100644 index 0000000000..c68fdb4b24 --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bias_bnorm_clamp.cpp @@ -0,0 +1,1272 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +using InLayout = ck::tensor_layout::convolution::NHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKYXC; +using OutLayout = ck::tensor_layout::convolution::NHWGK; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::BiasNormalizeInInferClamp; +using ck::tensor_operation::element_wise::PassThrough; + +template +using Quintet = ck::Tuple; + +template +struct DeviceOpHelper; + +template +struct DeviceOpHelper<2, T, ComputeType> +{ + using InLayout = ck::tensor_layout::convolution::NHWGC; + using WeiLayout = ck::tensor_layout::convolution::GKYXC; + using OutLayout = ck::tensor_layout::convolution::NHWGK; + + using Type = DeviceGroupedConvFwdMultipleABD<2, + InLayout, + WeiLayout, + Quintet, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + Quintet, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + BiasNormalizeInInferClamp, + ComputeType>; +}; + +template +struct DeviceOpHelper<3, T, ComputeType> +{ + using InLayout = ck::tensor_layout::convolution::NDHWGC; + using WeiLayout = ck::tensor_layout::convolution::GKZYXC; + using OutLayout = ck::tensor_layout::convolution::NDHWGK; + + using Type = DeviceGroupedConvFwdMultipleABD<3, + InLayout, + WeiLayout, + Quintet, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + Quintet, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + BiasNormalizeInInferClamp, + ComputeType>; +}; + +template +using DeviceOp = DeviceOpHelper::Type; + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F32_2D +{ + using DeviceOp = ::DeviceOp<2, float>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F32_TF32_2D +{ + using DeviceOp = ::DeviceOp<2, float, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F16_2D +{ + using DeviceOp = ::DeviceOp<2, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_2D +{ + using DeviceOp = ::DeviceOp<2, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK,NHWGK,NHWGK,NHWGK,NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +struct F32_3D +{ + using DeviceOp = ::DeviceOp<3, float>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F32_TF32_3D +{ + using DeviceOp = ::DeviceOp<3, float, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32,fp32,fp32,fp32,fp32),fp32,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F16_3D +{ + using DeviceOp = ::DeviceOp<3, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16,fp16,fp16,fp16,fp16),fp16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_3D +{ + using DeviceOp = ::DeviceOp<3, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK,NDHWGK,NDHWGK,NDHWGK,NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16,bf16,bf16,bf16,bf16),bf16,PassThrough,PassThrough,BiasNormalizeInInferClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +using TestTypes = + ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bias_clamp.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bias_clamp.cpp new file mode 100644 index 0000000000..a9e5e39e8f --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bias_clamp.cpp @@ -0,0 +1,1269 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +using InLayout = ck::tensor_layout::convolution::NHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKYXC; +using OutLayout = ck::tensor_layout::convolution::NHWGK; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::AddClamp; +using ck::tensor_operation::element_wise::PassThrough; + +template +struct DeviceOpHelper; + +template +struct DeviceOpHelper<2, T, ComputeType> +{ + using InLayout = ck::tensor_layout::convolution::NHWGC; + using WeiLayout = ck::tensor_layout::convolution::GKYXC; + using OutLayout = ck::tensor_layout::convolution::NHWGK; + + using Type = DeviceGroupedConvFwdMultipleABD<2, + InLayout, + WeiLayout, + ck::Tuple, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + ck::Tuple, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + AddClamp, + ComputeType>; +}; + +template +struct DeviceOpHelper<3, T, ComputeType> +{ + using InLayout = ck::tensor_layout::convolution::NDHWGC; + using WeiLayout = ck::tensor_layout::convolution::GKZYXC; + using OutLayout = ck::tensor_layout::convolution::NDHWGK; + + using Type = DeviceGroupedConvFwdMultipleABD<3, + InLayout, + WeiLayout, + ck::Tuple, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + ck::Tuple, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + AddClamp, + ComputeType>; +}; + +template +using DeviceOp = DeviceOpHelper::Type; + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F32_2D +{ + using DeviceOp = ::DeviceOp<2, float>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F32_TF32_2D +{ + using DeviceOp = ::DeviceOp<2, float, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F16_2D +{ + using DeviceOp = ::DeviceOp<2, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_2D +{ + using DeviceOp = ::DeviceOp<2, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,Tuple(NHWGK),NHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +struct F32_3D +{ + using DeviceOp = ::DeviceOp<3, float>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F32_TF32_3D +{ + using DeviceOp = ::DeviceOp<3, float, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp32,fp32,fp32,fp32,Tuple(fp32),fp32,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F16_3D +{ + using DeviceOp = ::DeviceOp<3, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,fp16,fp16,fp32,fp16,Tuple(fp16),fp16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_3D +{ + using DeviceOp = ::DeviceOp<3, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,Tuple(NDHWGK),NDHWGK,bf16,bf16,fp32,bf16,Tuple(bf16),bf16,PassThrough,PassThrough,AddClamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +using TestTypes = + ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bilinear.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bilinear.cpp new file mode 100644 index 0000000000..cc9cbd3dda --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_bilinear.cpp @@ -0,0 +1,118 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include +#include "ck/utility/data_type.hpp" +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +constexpr static auto NumDimSpatial = 3; +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; +using DsLayout = ck::Tuple; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::Bilinear; +using ck::tensor_operation::element_wise::PassThrough; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD, + type, // OutDataType + PassThrough, + PassThrough, + Bilinear, + computeType, + computeType>; + +} // namespace + +template +struct CkFactoryTestBilinearFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct Bilinear_F32 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct Bilinear_F32_TF32 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct Bilinear_F16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct Bilinear_BF16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct Bilinear_INT8 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +using TestTypes = + ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestBilinearFwd, TestTypes); + +TYPED_TEST(CkFactoryTestBilinearFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_clamp.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_clamp.cpp new file mode 100644 index 0000000000..a8d1d1763f --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_clamp.cpp @@ -0,0 +1,1249 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include +#include "ck/utility/data_type.hpp" +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; +using DsLayout = ck::Tuple<>; + +using ck::tensor_layout::convolution::GKYXC; +using ck::tensor_layout::convolution::GKZYXC; +using ck::tensor_layout::convolution::NDHWGC; +using ck::tensor_layout::convolution::NDHWGK; +using ck::tensor_layout::convolution::NHWGC; +using ck::tensor_layout::convolution::NHWGK; + +using DsLayout = ck::Tuple<>; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::Clamp; +using ck::tensor_operation::element_wise::PassThrough; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + Clamp, + ComputeType>; + +} // namespace + +template +struct CkFactoryTestBilinearFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F32_2D +{ + using DeviceOp = ::DeviceOp<2, float, NHWGC, GKYXC, NHWGK>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F32_3D +{ + using DeviceOp = ::DeviceOp<3, float, NDHWGC, GKZYXC, NDHWGK>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp32,fp32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp32,fp32>" + // clang-format on + }; +}; + +struct F32_TF32_2D +{ + using DeviceOp = ::DeviceOp<2, float, NHWGC, GKYXC, NHWGK, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F32_TF32_3D +{ + using DeviceOp = ::DeviceOp<3, float, NDHWGC, GKZYXC, NDHWGK, ck::tf32_t>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Interwave,v2,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,tf32,tf32>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp32,fp32,fp32,fp32,EmptyTuple,fp32,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,4,4,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,tf32,tf32>" + // clang-format on + }; +}; + +struct F16_2D +{ + using DeviceOp = ::DeviceOp<2, ck::half_t, NHWGC, GKYXC, NHWGK>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct F16_3D +{ + using DeviceOp = ::DeviceOp<3, ck::half_t, NDHWGC, GKZYXC, NDHWGK>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,fp16,fp16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,fp16,fp16,fp32,fp16,EmptyTuple,fp16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>" + // clang-format on + }; +}; + +struct BF16_2D +{ + using DeviceOp = ::DeviceOp<2, ck::bhalf_t, NHWGC, GKYXC, NHWGK>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<2,NHWGC,GKYXC,EmptyTuple,NHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +struct BF16_3D +{ + using DeviceOp = ::DeviceOp<3, ck::bhalf_t, NDHWGC, GKZYXC, NDHWGK>; + + constexpr static auto expected = { + // clang-format off + "", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Default,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,16,64,8,8,16,16,4,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,128,32,64,8,8,32,32,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,128,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,32,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,16,64,64,8,8,16,16,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,128,64,8,8,32,32,1,2,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,16,64,8,8,16,16,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,32,64,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,16,64,8,8,16,16,2,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,128,64,32,64,8,8,32,32,1,1,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,32,8,8,32,32,2,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,128,64,8,8,32,32,2,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,256,32,8,8,32,32,2,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,128,64,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,16,256,64,8,8,16,16,1,4,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,224,256,64,8,8,16,16,7,8,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,128,32,8,8,32,32,4,2,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,16,64,8,8,16,16,4,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,16,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),2,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,224,64,8,8,16,16,8,7,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,2,1,Seq(1,64,1,4),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,16,16,8,8,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,2,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v4,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,256,32,8,8,32,32,4,4,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(4,64,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v5,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,256,32,64,8,8,32,32,2,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,32,256,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,16),8,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,64,128,64,8,8,32,32,1,2,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,256,64,64,64,8,8,32,32,1,1,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,32,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,32,1,8),8,Intrawave,v3,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,128,8,8,16,16,1,1,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(16,4,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Interwave,v2,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v1,bf16,bf16>", + "DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle_V3<3,NDHWGC,GKZYXC,EmptyTuple,NDHWGK,bf16,bf16,fp32,bf16,EmptyTuple,bf16,PassThrough,PassThrough,Clamp,Filter1x1Stride1Pad0,MNKPadding,64,16,16,64,8,8,16,16,1,1,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2,8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,bf16,bf16>" + // clang-format on + }; +}; + +using TestTypes = + ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestBilinearFwd, TestTypes); + +TYPED_TEST(CkFactoryTestBilinearFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_convscale.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_convscale.cpp new file mode 100644 index 0000000000..0da300e3d8 --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_convscale.cpp @@ -0,0 +1,246 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include +#include +#include +#include +#include +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +constexpr static auto NumDimSpatial = 3; +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::device::instance::CombConvScale; +using ck::tensor_operation::device::instance::CombConvScaleRelu; +using ck::tensor_operation::element_wise::ConvInvscale; +using ck::tensor_operation::element_wise::ConvScale; +using ck::tensor_operation::element_wise::ConvScaleAdd; +using ck::tensor_operation::element_wise::ConvScaleRelu; +using ck::tensor_operation::element_wise::PassThrough; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD; + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F8_ConvScale +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::f8_t, + ck::f8_t, + ck::f8_t, + ConvScale, + ck::f8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_BF8_comb1_ConvScale +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::bf8_t, + ck::bf8_t, + ck::f8_t, + ConvScale, + ck::bf8_t, + ck::bf8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_BF8_comb2_ConvScale +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::f8_t, + ck::bf8_t, + ck::f8_t, + ConvScale, + ck::f8_t, + ck::bf8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_BF8_comb3_ConvScale +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::bf8_t, + ck::f8_t, + ck::f8_t, + ConvScale, + ck::bf8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_float_CombConvScale +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::f8_t, + ck::f8_t, + float, + CombConvScale, + ck::f8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_ConvScaleRelu +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::f8_t, + ck::f8_t, + ck::f8_t, + ConvScaleRelu, + ck::f8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_CombConvScaleRelu +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::f8_t, + ck::f8_t, + float, + CombConvScaleRelu, + ck::f8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_ConvScaleAdd +{ + using DeviceOp = ::DeviceOp, + ck::Tuple, + ck::f8_t, + ck::f8_t, + ck::f8_t, + ConvScaleAdd, + ck::f8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F8_ConvInvscale +{ + using DeviceOp = ::DeviceOp, + ck::Tuple<>, + ck::f8_t, + ck::f8_t, + ck::f8_t, + ConvInvscale, + ck::f8_t, + ck::f8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +using TestTypes = ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_dynamic_op.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_dynamic_op.cpp new file mode 100644 index 0000000000..e918785fd7 --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_dynamic_op.cpp @@ -0,0 +1,187 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include +#include "ck/utility/data_type.hpp" +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::DynamicUnaryOp; +using ck::tensor_operation::element_wise::PassThrough; + +template +struct DeviceOpHelper; + +template +struct DeviceOpHelper<2, T> +{ + using InLayout = ck::tensor_layout::convolution::NHWGC; + using WeiLayout = ck::tensor_layout::convolution::GKYXC; + using OutLayout = ck::tensor_layout::convolution::NHWGK; + + using Type = DeviceGroupedConvFwdMultipleABD<2, + InLayout, + WeiLayout, + ck::Tuple<>, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + ck::Tuple<>, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + DynamicUnaryOp>; +}; + +template +struct DeviceOpHelper<3, T> +{ + using InLayout = ck::tensor_layout::convolution::NDHWGC; + using WeiLayout = ck::tensor_layout::convolution::GKZYXC; + using OutLayout = ck::tensor_layout::convolution::NDHWGK; + + using Type = DeviceGroupedConvFwdMultipleABD<3, + InLayout, + WeiLayout, + ck::Tuple<>, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + ck::Tuple<>, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + DynamicUnaryOp>; +}; + +template +using DeviceOp = DeviceOpHelper::Type; + +} // namespace + +template +struct CkFactoryTestBilinearFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct DyOp_F32_2 +{ + using DeviceOp = ::DeviceOp<2, float>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_F32_3 +{ + using DeviceOp = ::DeviceOp<3, float>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_F16_2 +{ + using DeviceOp = ::DeviceOp<2, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_F16_3 +{ + using DeviceOp = ::DeviceOp<3, ck::half_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_BF16_2 +{ + using DeviceOp = ::DeviceOp<2, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_BF16_3 +{ + using DeviceOp = ::DeviceOp<3, ck::bhalf_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_INT8_2 +{ + using DeviceOp = ::DeviceOp<2, int8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct DyOp_INT8_3 +{ + using DeviceOp = ::DeviceOp<3, int8_t>; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +using TestTypes = ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestBilinearFwd, TestTypes); + +TYPED_TEST(CkFactoryTestBilinearFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scale.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scale.cpp new file mode 100644 index 0000000000..428d1c81f3 --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scale.cpp @@ -0,0 +1,115 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +constexpr static auto NumDimSpatial = 3; +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::PassThrough; +using ck::tensor_operation::element_wise::Scale; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + ck::Tuple<>, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + Scale, + ComputeType>; + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F32 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F32_TF32 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +using TestTypes = ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scaleadd_ab.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scaleadd_ab.cpp new file mode 100644 index 0000000000..774c30c05e --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scaleadd_ab.cpp @@ -0,0 +1,104 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +constexpr static auto NumDimSpatial = 3; +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; + +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::PassThrough; +using ck::tensor_operation::element_wise::ScaleAdd; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD, // DsLayout + OutLayout, + ck::Tuple, // InDataType + ck::Tuple, // WeiDataType + ck::Tuple<>, // DsDataType + T, // OutDataType + ScaleAdd, + ScaleAdd, + PassThrough, + T>; // ComputeType + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F32 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +using TestTypes = ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scaleadd_scaleadd_relu.cpp b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scaleadd_scaleadd_relu.cpp new file mode 100644 index 0000000000..ba8726a643 --- /dev/null +++ b/experimental/builder/test/test_ck_factory_grouped_convolution_forward_scaleadd_scaleadd_relu.cpp @@ -0,0 +1,105 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceSet; +using ck_tile::test::InstancesMatch; + +namespace { + +constexpr static auto NumDimSpatial = 3; +using InLayout = ck::tensor_layout::convolution::NDHWGC; +using WeiLayout = ck::tensor_layout::convolution::GKZYXC; +using OutLayout = ck::tensor_layout::convolution::NDHWGK; + +using ck::tensor_layout::convolution::G_K; +using ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD; +using ck::tensor_operation::element_wise::PassThrough; +using ck::tensor_operation::element_wise::ScaleAddScaleAddRelu; + +template +using DeviceOp = DeviceGroupedConvFwdMultipleABD, // DsLayout + OutLayout, + T, // InDataType + T, // WeiDataType + ck::Tuple, // DsDataType + T, // OutDataType + PassThrough, + PassThrough, + ScaleAddScaleAddRelu, + T>; // ComputeType + +} // namespace + +template +struct CkFactoryTestConvFwd : public testing::Test +{ + static auto get_actual_instances() + { + return InstanceSet::from_factory(); + } + + static auto get_expected_instances() { return InstanceSet(Case::expected); } +}; + +struct F32 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct F16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct BF16 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +struct S8 +{ + using DeviceOp = ::DeviceOp; + + constexpr static auto expected = { + // clang-format off + "" + // clang-format on + }; +}; + +using TestTypes = ::testing::Types; + +TYPED_TEST_SUITE(CkFactoryTestConvFwd, TestTypes); + +TYPED_TEST(CkFactoryTestConvFwd, TestInstances) +{ + auto actual = TestFixture::get_actual_instances(); + auto expected = TestFixture::get_expected_instances(); + + EXPECT_THAT(actual, InstancesMatch(expected)); +} diff --git a/experimental/builder/test/test_testing_utils.cpp b/experimental/builder/test/test_testing_utils.cpp new file mode 100644 index 0000000000..24a1c9bc81 --- /dev/null +++ b/experimental/builder/test/test_testing_utils.cpp @@ -0,0 +1,98 @@ +// SPDX-License-Identifier: MIT +// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved. + +#include + +#include "testing_utils.hpp" + +using ck_tile::test::InstanceMatcher; +using ck_tile::test::InstanceSet; +using ck_tile::test::StringEqWithDiff; + +TEST(InstanceSet, FromFactory) +{ + using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD< + 2, // NDimSpatial + ck::tensor_operation::device::instance::NHWGC, // InLayout + ck::tensor_operation::device::instance::GKYXC, // WeiLayout + ck::tensor_operation::device::instance::Empty_Tuple, // DsLayout + ck::tensor_operation::device::instance::NHWGK, // OutLayout + ck::half_t, // ADataType + ck::half_t, // BDataType + ck::Tuple<>, // DsDataType + ck::half_t, // EDataType + ck::tensor_operation::element_wise::PassThrough, // AElementwiseOperation + ck::tensor_operation::element_wise::PassThrough, // BElementwiseOperation + ck::tensor_operation::element_wise::PassThrough, // CDEElementwiseOperation + ck::half_t, // AComputeType + ck::half_t>; // BComputeType + + const auto instances = InstanceSet::from_factory(); + + EXPECT_THAT(instances.instances.size(), testing::Gt(0)); + + const auto* el = + "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),2,8,8,0,Seq(8,8,1),Seq(1,0,2),Seq(1,0,2),2," + "8,8,0,1,1,Seq(1,16,1,4),4,Intrawave,v2,fp16,fp16>"; + EXPECT_THAT(instances.instances, testing::Contains(el)); +} + +TEST(InstanceMatcher, Basic) +{ + auto a = InstanceSet{ + "python", + "cobra", + "boa", + }; + + auto b = InstanceSet{ + "cobra", + "boa", + "python", + }; + + auto c = InstanceSet{ + "adder", + "boa", + "cobra", + }; + + auto d = InstanceSet{ + "boa", + "python", + }; + + EXPECT_THAT(a, InstancesMatch(b)); + EXPECT_THAT(c, Not(InstancesMatch(b))); + EXPECT_THAT(a, Not(InstancesMatch(d))); + EXPECT_THAT(d, Not(InstancesMatch(b))); +} + +TEST(InstanceMatcher, ExplainMatchResult) +{ + auto actual = InstanceSet{ + "python", + "cobra", + "boa", + }; + + auto expected = InstanceSet{ + "adder", + "boa", + "cobra", + "rattlesnake", + }; + + testing::StringMatchResultListener listener; + EXPECT_TRUE(!ExplainMatchResult(InstancesMatch(expected), actual, &listener)); + + EXPECT_THAT(listener.str(), + StringEqWithDiff("\n" + " Missing: 2\n" + "- adder\n" + "- rattlesnake\n" + "Unexpected: 1\n" + "- python\n")); +} diff --git a/experimental/builder/test/testing_utils.cpp b/experimental/builder/test/testing_utils.cpp index c99d56ef56..34793b601e 100644 --- a/experimental/builder/test/testing_utils.cpp +++ b/experimental/builder/test/testing_utils.cpp @@ -1,21 +1,18 @@ // SPDX-License-Identifier: MIT // Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. -#include -#include -#include -#include -#include +#include "testing_utils.hpp" #include #include -#include "testing_utils.hpp" +#include +#include +#include +#include +#include +#include namespace ck_tile::test { -namespace { - -} // namespace - // Wagner-Fischer Algorithm for Computing Edit Distance and Inline Diff // // OUTPUT FORMAT: [expected|actual] for differences, plain text for matches @@ -216,4 +213,88 @@ void StringEqWithDiffMatcher::DescribeNegationTo(std::ostream* os) const return ::testing::MakeMatcher(new StringEqWithDiffMatcher(expected)); } +std::ostream& operator<<(std::ostream& os, const InstanceSet& set) +{ + // These sets can grow very large, and so its not very nice or useful to print them + // in the event of a mismatch. Just print a brief description here, and use + // InstancesMatcher to print a more useful message. + return (os << "(set of " << set.instances.size() << " instances)"); +} + +InstanceMatcher::InstanceMatcher(const InstanceSet& expected) : expected_(expected) {} + +::testing::Matcher InstancesMatch(const InstanceSet& expected) +{ + return ::testing::MakeMatcher(new InstanceMatcher(expected)); +} + +bool InstanceMatcher::MatchAndExplain(InstanceSet actual, + ::testing::MatchResultListener* listener) const +{ + if(actual.instances == expected_.instances) + { + return true; + } + + if(listener->IsInterested()) + { + std::vector instances; + std::set_difference(expected_.instances.begin(), + expected_.instances.end(), + actual.instances.begin(), + actual.instances.end(), + std::back_inserter(instances)); + + *listener << "\n"; + + if(instances.size() > 0) + { + *listener << " Missing: " << instances.size() << "\n"; + for(const auto& instance : instances) + { + if(instance == "") + { + *listener << "- (empty string)\n"; + } + else + { + *listener << "- " << instance << "\n"; + } + } + } + + instances.clear(); + std::set_difference(actual.instances.begin(), + actual.instances.end(), + expected_.instances.begin(), + expected_.instances.end(), + std::back_inserter(instances)); + + if(instances.size() > 0) + { + *listener << "Unexpected: " << instances.size() << "\n"; + for(const auto& instance : instances) + { + if(instance == "") + { + *listener << "- (empty string)\n"; + } + else + { + *listener << "- " << instance << "\n"; + } + } + } + } + + return false; +} + +void InstanceMatcher::DescribeTo(std::ostream* os) const { *os << expected_; } + +void InstanceMatcher::DescribeNegationTo(std::ostream* os) const +{ + *os << "is not equal to " << expected_; +} + } // namespace ck_tile::test diff --git a/experimental/builder/test/testing_utils.hpp b/experimental/builder/test/testing_utils.hpp index 3e8772a080..3ff2eb32de 100644 --- a/experimental/builder/test/testing_utils.hpp +++ b/experimental/builder/test/testing_utils.hpp @@ -1,10 +1,15 @@ // SPDX-License-Identifier: MIT // Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved. +#include #include #include #include #include +#include +#include +#include +#include namespace ck_tile::test { @@ -40,4 +45,68 @@ class StringEqWithDiffMatcher : public ::testing::MatcherInterface // Factory function for the StringEqWithDiff matcher ::testing::Matcher StringEqWithDiff(const std::string& expected); +using ck::tensor_operation::device::instance::DeviceOperationInstanceFactory; + +// This utility concept checks whether a type is a valid "Device Operation" - +// that is, there is a valid specialization of `DeviceOperationInstanceFactory` +// for it available. +template +concept HasCkFactory = requires { + { + DeviceOperationInstanceFactory::GetInstances() + } -> std::convertible_to>>; +}; + +// This structure represents a (unique) set of instances, either a statically +// defined one (for testing) or one obtained from DeviceOperationInstanceFactory. +// The idea is that we use this structure as a utility to compare a set of +// instances. Instances are stored in a set so that they can be lexicographically +// compared, this helps generating readable error messages which just contain +// the differenses between sets. +struct InstanceSet +{ + explicit InstanceSet() {} + + explicit InstanceSet(std::initializer_list items) + : instances(items.begin(), items.end()) + { + } + + template + static InstanceSet from_factory() + { + auto set = InstanceSet(); + + const auto ops = DeviceOperationInstanceFactory::GetInstances(); + for(const auto& op : ops) + { + set.instances.insert(op->GetInstanceString()); + } + + return set; + } + + std::set instances; +}; + +std::ostream& operator<<(std::ostream& os, const InstanceSet& set); + +// This is a custom Google Test matcher which can be used to compare two sets +// of instance names, with utility functions that print a helpful error +// message about the difference between the checked sets. Use `InstancesMatch` +// to obtain an instance of this type. +struct InstanceMatcher : public ::testing::MatcherInterface +{ + explicit InstanceMatcher(const InstanceSet& expected); + + bool MatchAndExplain(InstanceSet actual, + ::testing::MatchResultListener* listener) const override; + void DescribeTo(std::ostream* os) const override; + void DescribeNegationTo(std::ostream* os) const override; + + InstanceSet expected_; +}; + +::testing::Matcher InstancesMatch(const InstanceSet& expected); + } // namespace ck_tile::test diff --git a/include/ck_tile/ops/topk_softmax/kernel/topk_softmax_kernel.hpp b/include/ck_tile/ops/topk_softmax/kernel/topk_softmax_kernel.hpp index e8727ea065..019e940a33 100644 --- a/include/ck_tile/ops/topk_softmax/kernel/topk_softmax_kernel.hpp +++ b/include/ck_tile/ops/topk_softmax/kernel/topk_softmax_kernel.hpp @@ -21,7 +21,7 @@ struct TopkSoftmaxHostArgs index_t num_experts; index_t topk; index_t stride_input; // row stride for input, at least experts - index_t stride_output; // row stride for output/indices, at least tpok + index_t stride_output; // row stride for output/indices, at least topk }; template @@ -45,7 +45,7 @@ struct TopkSoftmaxKernel index_t num_experts; index_t topk; index_t stride_input; // row stride for input, at least experts - index_t stride_output; // row stride for output/indices, at least tpok + index_t stride_output; // row stride for output/indices, at least topk }; using Kargs = TopkSoftmaxKargs; diff --git a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_pipeline.hpp b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_pipeline.hpp index d620d9bec9..677263229b 100644 --- a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_pipeline.hpp +++ b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_pipeline.hpp @@ -90,6 +90,11 @@ struct TopkSoftmaxWarpPerRowPipeline const auto current_expert = x_indices.at(number<1>{}); w_(idx) = current_expert >= experts ? -numeric::infinity() : w_(idx); + if constexpr(!Problem::ActivationIsSoftmax) + { + // sigmoid can be pre-computed already here if not using softmax + w_(idx) = WeightType(1) / (WeightType(1) + exp(-w_(idx))); + } }; tile_sweeper ts{w_, w_f}; ts(); @@ -97,10 +102,16 @@ struct TopkSoftmaxWarpPerRowPipeline #endif }(); - // softmax - auto y = softmax(w); - - topk(y, out_win, idx_win, k); + if constexpr(Problem::ActivationIsSoftmax) + { + auto y = softmax(w); + topk(y, out_win, idx_win, k); + } + else + { + // sigmoid was already pre-computed above, so only do topk now + topk(w, out_win, idx_win, k); + } // check exit if constexpr(Problem::LaunchType == 0) diff --git a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp index 917096ad5e..1dc7e9335e 100644 --- a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp +++ b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp @@ -13,10 +13,11 @@ template 0, persistent #occupancy - index_t BlockSize_ = 256> + bool ActivationIsSoftmax_ = true, // false: sigmoid + index_t IssuesPerCol_ = 2, // issue along col, to make sure block_reduce() OK + index_t BytesPerIssue_ = sizeof(InputType_), + index_t LaunchType_ = 0, // 0-streaming, >0, persistent #occupancy + index_t BlockSize_ = 256> struct TopkSoftmaxWarpPerRowProblem { // TODO: this kernel only support warp per row @@ -31,6 +32,8 @@ struct TopkSoftmaxWarpPerRowProblem static constexpr index_t BlockSize = BlockSize_; static constexpr index_t WarpSize = get_warp_size(); + static constexpr bool ActivationIsSoftmax = ActivationIsSoftmax_; + static_assert(BytesPerIssue % sizeof(InputType) == 0); static constexpr index_t VectorSize = BytesPerIssue / sizeof(InputType); static_assert(Experts % VectorSize == 0); diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp index ea871f3325..b086a8fbab 100644 --- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp +++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp @@ -300,7 +300,7 @@ struct DeviceOperationInstanceFactory && is_same_v && is_same_v) { diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp index 1c873863b3..588bb554ad 100644 --- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp +++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp @@ -7,7 +7,7 @@ #include #include "ck/ck.hpp" -#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dynamic.hpp" +#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_abd.hpp" #include "ck/tensor_operation/gpu/device/tensor_layout.hpp" #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp" @@ -161,7 +161,7 @@ template + typename BComputeType> struct DeviceOperationInstanceFactory>>& instances); -void add_device_grouped_conv3d_fwd_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_instances( +void add_device_grouped_conv3d_fwd_xdl_scale_ndhwgc_gkzyxc_ndhwgk_f32_tf32_instances( std::vector& x, reference_topk(y, y_values, y_indices, k, dim, largest, sorted); } +template +auto reference_topk_sigmoid(const ck_tile::HostTensor& x, + ck_tile::HostTensor& y_values, + ck_tile::HostTensor& y_indices, + ck_tile::index_t k, + ck_tile::index_t dim = -1, + bool largest = true, + bool sorted = true) +{ + using namespace ck_tile; + + // topk only - no need to apply the sigmoid first + auto x_fp32 = x.template CopyAsType(); + reference_topk(x_fp32, y_values, y_indices, k, dim, largest, sorted); + // apply sigmoid + std::transform(y_values.begin(), y_values.end(), y_values.begin(), [](auto value) { + return WeightType(1) / (WeightType(1) + exp(-value)); + }); +} + // different threshold for different dtype template auto get_elimit(std::string /*init_method*/) @@ -87,7 +107,8 @@ auto create_args(int argc, char* argv[]) .insert("seed", "-1", "seed to be used, -1 means random every time") .insert("kname", "0", "when set to 1 it will print kernel name") .insert("warmup", "5", "number of iterations before benchmark the kernel") - .insert("repeat", "20", "number of iterations to benchmark the kernel"); + .insert("repeat", "20", "number of iterations to benchmark the kernel") + .insert("activation", "softmax", "activation function to use: softmax or sigmoid"); bool result = arg_parser.parse(argc, argv); return std::make_tuple(result, arg_parser); @@ -108,6 +129,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) int kname = args.get_int("kname"); int warmup = args.get_int("warmup"); int repeat = args.get_int("repeat"); + std::string activation = args.get_str("activation"); if(stride_input < 0) { @@ -158,7 +180,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) x_dev.ToDevice(x_host.data()); - topk_softmax_trait trait{input_prec, weight_prec, experts}; + topk_softmax_trait trait{input_prec, weight_prec, experts, activation}; topk_softmax_kargs karg{x_dev.GetDeviceBuffer(), value_dev.GetDeviceBuffer(), @@ -175,7 +197,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) warmup, repeat}; auto ms = topk_softmax(trait, karg, sc); - printf("[%s|%s]tokens:%d, experts:%d, topk:%d, st_i:%d, st_o:%d, ms:%f, ", + printf("[%s|%s]tokens:%d, experts:%d, topk:%d, st_i:%d, st_o:%d, activation:%s, ms:%f, ", input_prec.c_str(), weight_prec.c_str(), tokens, @@ -183,6 +205,7 @@ bool test_topk_softmax(ck_tile::ArgParser args) topk, stride_input, stride_output, + activation.c_str(), ms); if(ms < 0) printf("not supported\n"); @@ -201,8 +224,20 @@ bool test_topk_softmax(ck_tile::ArgParser args) ck_tile::HostTensor value_ref({tokens, topk}, {stride_output, 1}); ck_tile::HostTensor index_ref({tokens, topk}, {stride_output, 1}); - reference_topk_softmax( - x_host, value_ref, index_ref, topk); + if(activation == "softmax") + { + reference_topk_softmax( + x_host, value_ref, index_ref, topk); + } + else if(activation == "sigmoid") + { + reference_topk_sigmoid( + x_host, value_ref, index_ref, topk); + } + else + { + throw std::runtime_error("unsupported activation type: " + activation); + } auto [rtol, atol] = get_elimit(""); for(int i_t = 0; i_t < tokens; i_t++) @@ -255,7 +290,10 @@ int run_gemm_combinations(std::string const& data_type) {"-t=71", "-e=11", "-k=11", "-st_i=30", "-st_o=12"}, {"-t=1", "-e=1", "-k=1"}, {"-t=99", "-e=2", "-k=1", "-st_i=11", "-st_o=5"}, - {"-t=333", "-e=99", "-k=13", "-st_i=191", "-st_o=17"}}; + {"-t=333", "-e=99", "-k=13", "-st_i=191", "-st_o=17"}, + {"-t=20", "-e=5", "-k=2", "-activation=sigmoid"}, + {"-t=220", "-e=9", "-k=3", "-activation=sigmoid"}, + {"-t=500", "-e=21", "-k=13", "-activation=sigmoid"}}; bool result = true; std::string pr_i = "-pr_i=" + data_type; diff --git a/test/ck_tile/topk_softmax/test_topk_softmax_api.cpp b/test/ck_tile/topk_softmax/test_topk_softmax_api.cpp index 7c90c8200c..7e9ed6301a 100644 --- a/test/ck_tile/topk_softmax/test_topk_softmax_api.cpp +++ b/test/ck_tile/topk_softmax/test_topk_softmax_api.cpp @@ -3,27 +3,31 @@ #include "test_topk_softmax_api.hpp" -#define TOPK_SOFTMAX_DISPATCH(experts_) \ - constexpr ck_tile::index_t ts_experts = experts_; \ - using ts_problem = ck_tile:: \ - TopkSoftmaxWarpPerRowProblem; \ - using ts_pipeline = ck_tile::TopkSoftmaxWarpPerRowPipeline; \ - \ - using kernel = ck_tile::TopkSoftmaxKernel; \ - \ - auto kargs = kernel::MakeKargs(a); \ - \ - const dim3 grids = kernel::GridSize(a); \ - const dim3 blocks = kernel::BlockSize(); \ - \ - float ave_time = \ - ck_tile::launch_kernel(s, ck_tile::make_kernel<1>(kernel{}, grids, blocks, 0, kargs)); \ - \ +#define TOPK_SOFTMAX_DISPATCH(experts_, use_softmax_) \ + constexpr ck_tile::index_t ts_experts = experts_; \ + constexpr bool ts_use_softmax = use_softmax_; \ + using ts_problem = ck_tile::TopkSoftmaxWarpPerRowProblem; \ + using ts_pipeline = ck_tile::TopkSoftmaxWarpPerRowPipeline; \ + \ + using kernel = ck_tile::TopkSoftmaxKernel; \ + \ + auto kargs = kernel::MakeKargs(a); \ + \ + const dim3 grids = kernel::GridSize(a); \ + const dim3 blocks = kernel::BlockSize(); \ + \ + float ave_time = \ + ck_tile::launch_kernel(s, ck_tile::make_kernel<1>(kernel{}, grids, blocks, 0, kargs)); \ + \ return ave_time; float topk_softmax(topk_softmax_trait t, topk_softmax_kargs a, ck_tile::stream_config s) { - if(t.input_type == "fp16" && t.weight_type == "fp32") + if(t.input_type == "fp16" && t.weight_type == "fp32" && t.activation == "softmax") { using ts_input_type = ck_tile::fp16_t; using ts_weight_type = float; @@ -31,36 +35,36 @@ float topk_softmax(topk_softmax_trait t, topk_softmax_kargs a, ck_tile::stream_c #if 1 if(t.experts <= 8) { - TOPK_SOFTMAX_DISPATCH(8) + TOPK_SOFTMAX_DISPATCH(8, true) } else if(t.experts <= 16) { - TOPK_SOFTMAX_DISPATCH(16) + TOPK_SOFTMAX_DISPATCH(16, true) } else if(t.experts <= 32) { - TOPK_SOFTMAX_DISPATCH(32) + TOPK_SOFTMAX_DISPATCH(32, true) } else if(t.experts <= 64) { - TOPK_SOFTMAX_DISPATCH(64) + TOPK_SOFTMAX_DISPATCH(64, true) } else if(t.experts <= 128) { - TOPK_SOFTMAX_DISPATCH(128) + TOPK_SOFTMAX_DISPATCH(128, true) } else if(t.experts <= 192) { - TOPK_SOFTMAX_DISPATCH(192) + TOPK_SOFTMAX_DISPATCH(192, true) } #else if(t.experts <= 128) { - TOPK_SOFTMAX_DISPATCH(128) + TOPK_SOFTMAX_DISPATCH(128, true) } #endif } - else if(t.input_type == "bf16" && t.weight_type == "fp32") + else if(t.input_type == "bf16" && t.weight_type == "fp32" && t.activation == "softmax") { #if 1 using ts_input_type = ck_tile::bf16_t; @@ -68,27 +72,96 @@ float topk_softmax(topk_softmax_trait t, topk_softmax_kargs a, ck_tile::stream_c using ts_index_type = ck_tile::index_t; if(t.experts <= 8) { - TOPK_SOFTMAX_DISPATCH(8) + TOPK_SOFTMAX_DISPATCH(8, true) } else if(t.experts <= 16) { - TOPK_SOFTMAX_DISPATCH(16) + TOPK_SOFTMAX_DISPATCH(16, true) } else if(t.experts <= 32) { - TOPK_SOFTMAX_DISPATCH(32) + TOPK_SOFTMAX_DISPATCH(32, true) } else if(t.experts <= 64) { - TOPK_SOFTMAX_DISPATCH(64) + TOPK_SOFTMAX_DISPATCH(64, true) } else if(t.experts <= 128) { - TOPK_SOFTMAX_DISPATCH(128) + TOPK_SOFTMAX_DISPATCH(128, true) } else if(t.experts <= 192) { - TOPK_SOFTMAX_DISPATCH(192) + TOPK_SOFTMAX_DISPATCH(192, true) + } +#endif + } + else if(t.input_type == "fp16" && t.weight_type == "fp32" && t.activation == "sigmoid") + { + using ts_input_type = ck_tile::fp16_t; + using ts_weight_type = float; + using ts_index_type = ck_tile::index_t; +#if 1 + if(t.experts <= 8) + { + TOPK_SOFTMAX_DISPATCH(8, false) + } + else if(t.experts <= 16) + { + TOPK_SOFTMAX_DISPATCH(16, false) + } + else if(t.experts <= 32) + { + TOPK_SOFTMAX_DISPATCH(32, false) + } + else if(t.experts <= 64) + { + TOPK_SOFTMAX_DISPATCH(64, false) + } + else if(t.experts <= 128) + { + TOPK_SOFTMAX_DISPATCH(128, false) + } + else if(t.experts <= 192) + { + TOPK_SOFTMAX_DISPATCH(192, false) + } +#else + if(t.experts <= 128) + { + TOPK_SOFTMAX_DISPATCH(128, false) + } +#endif + } + else if(t.input_type == "bf16" && t.weight_type == "fp32" && t.activation == "sigmoid") + { +#if 1 + using ts_input_type = ck_tile::bf16_t; + using ts_weight_type = float; + using ts_index_type = ck_tile::index_t; + if(t.experts <= 8) + { + TOPK_SOFTMAX_DISPATCH(8, false) + } + else if(t.experts <= 16) + { + TOPK_SOFTMAX_DISPATCH(16, false) + } + else if(t.experts <= 32) + { + TOPK_SOFTMAX_DISPATCH(32, false) + } + else if(t.experts <= 64) + { + TOPK_SOFTMAX_DISPATCH(64, false) + } + else if(t.experts <= 128) + { + TOPK_SOFTMAX_DISPATCH(128, false) + } + else if(t.experts <= 192) + { + TOPK_SOFTMAX_DISPATCH(192, false) } #endif } diff --git a/test/ck_tile/topk_softmax/test_topk_softmax_api.hpp b/test/ck_tile/topk_softmax/test_topk_softmax_api.hpp index 65651efa4d..c98a887736 100644 --- a/test/ck_tile/topk_softmax/test_topk_softmax_api.hpp +++ b/test/ck_tile/topk_softmax/test_topk_softmax_api.hpp @@ -12,6 +12,7 @@ struct topk_softmax_trait std::string input_type; std::string weight_type; // currently always float int experts; + std::string activation; // "softmax" or "sigmoid" }; struct topk_softmax_kargs : public ck_tile::TopkSoftmaxHostArgs