diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_two_stage_xdl_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_two_stage_xdl_instance.hpp
index 1c4dc8a445..0ed12b984b 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_two_stage_xdl_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_two_stage_xdl_instance.hpp
@@ -72,7 +72,14 @@ using device_grouped_conv_bwd_weight_two_stage_nhwgc_xdl_c_shuffle_f16_instances
 
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    32,    32,     32,   8,   32,   32,    1,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              2,              2,      false,  S<4, 8,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              2,              2,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 2>,
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    64,    32,     32,   8,   32,   32,    2,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              4,              4,      false,  S<4, 8,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              4,              4,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 4>,
-        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 8>
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 8>,
+
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1,  S<8, 8, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              8,              8,      false,  S<8, 8,  1>,  S<2, 0, 1>,  S<2, 0, 1>,                1,              8,              8,      false,           1,           1,   S<1, 16, 1, 16>,                4, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 16, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              4,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              1,              8,      false,           1,           1,    S<1, 4, 1, 64>,                1, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              1,              8,      false,  S<8, 16,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              4,              8,      false,           1,           1,   S<1, 16, 1, 16>,                4, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              2,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              2,              8,      false,           1,           1,    S<1, 8, 1, 32>,                2, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              1,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              1,              8,      false,           1,           1,    S<1, 4, 1, 64>,                1, Scheduler, PipelineVersion, 1>
+
     // clang-format on
     >;
 
@@ -138,7 +145,13 @@ using device_grouped_conv_bwd_weight_two_stage_nhwgc_xdl_c_shuffle_bf16_instance
 
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    32,    32,     32,   8,   32,   32,    1,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              2,              2,      false,  S<4, 8,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              2,              2,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 2>,
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    64,    32,     32,   8,   32,   32,    2,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              4,              4,      false,  S<4, 8,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              4,              4,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 4>,
-        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 8>
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 8>,
+
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1,  S<8, 4, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              8,              8,      false,  S<8, 4,  1>,  S<2, 0, 1>,  S<2, 0, 1>,                1,              8,              8,      false,           1,           1,   S<1, 16, 1, 16>,                4, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 16, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              4,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              1,              8,      false,           1,           1,    S<1, 4, 1, 64>,                1, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              1,              8,      false,  S<8, 16,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              4,              8,      false,           1,           1,   S<1, 16, 1, 16>,                4, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              2,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              2,              8,      false,           1,           1,    S<1, 8, 1, 32>,                2, Scheduler, PipelineVersion, 1>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              1,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              1,              8,      false,           1,           1,    S<1, 4, 1, 64>,                1, Scheduler, PipelineVersion, 1>
     // clang-format on
     >;
 
@@ -218,7 +231,11 @@ using device_grouped_conv_bwd_weight_two_stage_ngchw_xdl_c_shuffle_f16_instances
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    32,   128,     32,   8,   32,   32,    1,    4,  S<4, 4,  1>, S<2, 0, 1>,  S<1, 0, 2>,                   1,              8,              8,      false,  S<4, 16,  1>, S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 4, 1, 8>,                  1, Scheduler, PipelineVersion, 8, F16, F16, 8 ,1>,
 
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    64,    32,     32,   8,   32,   32,    2,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              4,              4,      false,  S<4, 8,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              4,              4,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 4, F16, F16, 4, 1>,
-        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 4>,                  1, Scheduler, PipelineVersion, 8, F16, F16, 8, 1>
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 4>,                  1, Scheduler, PipelineVersion, 8, F16, F16, 8, 1>,
+
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1,  S<8, 8, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              8,              8,      false,  S<8, 8,  1>,  S<2, 0, 1>,  S<2, 0, 1>,                1,              8,              8,      false,           1,           1,   S<1, 16, 1, 16>,                4, Scheduler, PipelineVersion, 1, F16, F16, 4, 4>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              2,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              2,              8,      false,           1,           1,    S<1, 8, 1, 32>,                2, Scheduler, PipelineVersion, 1, F16, F16, 2, 2>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F16,     F16,     F16,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              1,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              1,              8,      false,           1,           1,    S<1, 4, 1, 64>,                1, Scheduler, PipelineVersion, 1, F16, F16, 1, 1>
     // clang-format on
     >;
 
@@ -275,7 +292,11 @@ using device_grouped_conv_bwd_weight_two_stage_ngchw_xdl_c_shuffle_bf16_instance
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    32,   128,     32,   8,   32,   32,    1,    4,  S<4, 4,  1>, S<2, 0, 1>,  S<1, 0, 2>,                   1,              8,              8,      false,  S<4, 16,  1>, S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 4, 1, 8>,                  1, Scheduler, PipelineVersion, 8, BF16, BF16, 8 ,1>,
 
         DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    64,    32,     32,   8,   32,   32,    2,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              4,              4,      false,  S<4, 8,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              4,              4,      false,           1,           1,   S<1, 8, 1, 8>,                  1, Scheduler, PipelineVersion, 4, BF16, BF16, 4, 1>,
-        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 4>,                  1, Scheduler, PipelineVersion, 8, BF16, BF16, 8, 1>
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,   128,    32,     32,   8,   32,   32,    4,    1,  S<4, 16, 1>,  S<2, 0, 1>,  S<1, 0, 2>,                  1,              8,              8,      false,  S<4, 4,  1>,  S<2, 0, 1>,  S<1, 0, 2>,                1,              8,              8,      false,           1,           1,   S<1, 8, 1, 4>,                  1, Scheduler, PipelineVersion, 8, BF16, BF16, 8, 1>,
+
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1,  S<8, 8, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              8,              8,      false,  S<8, 8,  1>,  S<2, 0, 1>,  S<2, 0, 1>,                1,              8,              8,      false,           1,           1,   S<1, 16, 1, 16>,                4, Scheduler, PipelineVersion, 1, BF16, BF16, 4, 4>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              2,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              2,              8,      false,           1,           1,    S<1, 8, 1, 32>,                2, Scheduler, PipelineVersion, 1, BF16, BF16, 2, 2>,
+        DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,   BF16,    BF16,    BF16,    F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     64,   8,   32,   32,    1,    1, S<8, 32, 1>,   S<2, 0, 1>,  S<2, 0, 1>,                  1,              1,              8,      false,  S<8, 32,  1>, S<2, 0, 1>,  S<2, 0, 1>,                1,              1,              8,      false,           1,           1,    S<1, 4, 1, 64>,                1, Scheduler, PipelineVersion, 1, BF16, BF16, 1, 1>
     // clang-format on
     >;
 
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_instance.hpp
index a493719637..3587570e42 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_instance.hpp
@@ -87,7 +87,12 @@ using device_grouped_conv_bwd_weight_xdl_c_shuffle_f32_instances = std::tuple<
         DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   128,   128,    32,     4,  4,   32,   32,    2,    1,  S<1, 4, 32, 1>,   S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              4,              4,      true,  S<1, 4, 8,  4>,  S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              4,              1,      true,           1,           1,   S<1, 32, 1, 4>,               4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
         DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   128,    32,   128,     4,  4,   32,   32,    1,    2,  S<1, 4, 8,  4>,   S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              4,              1,      true,  S<1, 4, 32, 1>,  S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              4,              4,      true,           1,           1,   S<1, 32, 1, 4>,               4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
         DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    64,    32,     4,  4,   32,   32,    2,    1,  S<1, 4, 16, 1>,   S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              4,              4,      true,  S<1, 4, 8,  2>,  S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              4,              2,      true,           1,           1,   S<1, 16, 1, 4>,               4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
-        DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    32,    64,     4,  4,   32,   32,    1,    2,  S<1, 4, 8,  2>,   S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              4,              2,      true,  S<1, 4, 16, 1>,  S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              4,              4,      true,           1,           1,   S<1, 16, 1, 4>,               4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>
+        DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,    64,    32,    64,     4,  4,   32,   32,    1,    2,  S<1, 4, 8,  2>,   S<0, 3, 1, 2>,   S<0, 2, 1, 3>,              2,              4,              2,      true,  S<1, 4, 16, 1>,  S<0, 3, 1, 2>,  S<0, 2, 1, 3>,             2,              4,              4,      true,           1,           1,   S<1, 16, 1, 4>,               4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
+
+        DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     8,  8,   32,   32,    1,    1, S<1, 8, 16, 1>,   S<0, 3, 1, 2>,  S<0, 3, 1, 2>,                2,              4,              4,      true,  S<1, 8, 16,  1>,  S<0, 3, 1, 2>,  S<0, 3, 1, 2>,             2,             4,              4,      true,           1,           1,   S<1, 16, 1, 16>,              4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
+        DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     8,  8,   32,   32,    1,    1, S<1, 8, 16, 1>,   S<0, 3, 1, 2>,  S<0, 3, 1, 2>,                2,              4,              4,      true,  S<1, 8, 32,  1>, S<0, 3, 1, 2>,  S<0, 3, 1, 2>,             2,              1,              4,      true,           1,           1,    S<1, 4, 1, 64>,              1, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
+        DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     8,  8,   32,   32,    1,    1, S<1, 8, 32, 1>,   S<0, 3, 1, 2>,  S<0, 3, 1, 2>,                2,              1,              4,      true,  S<1, 8, 16,  1>, S<0, 3, 1, 2>,  S<0, 3, 1, 2>,             2,              4,              4,      true,           1,           1,   S<1, 16, 1, 16>,              4, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>,
+        DeviceGroupedConvBwdWeight_Xdl_CShuffle< NDimSpatial,  ALayout,   BLayout,   ELayout,    F32,     F32,     F32,     F32, PassThrough, PassThrough, PassThrough,                  ConvSpec,   256,    64,    64,     8,  8,   32,   32,    1,    1, S<1, 8, 32, 1>,   S<0, 3, 1, 2>,  S<0, 3, 1, 2>,                2,              1,              4,      true,  S<1, 8, 32,  1>, S<0, 3, 1, 2>,  S<0, 3, 1, 2>,             2,              1,              4,      true,           1,           1,    S<1, 4, 1, 64>,              1, F32, F32, MaxTransposeTransferSrcScalarPerVector, MaxTransposeTransferDstScalarPerVector>
     // clang-format on
     >;