Merge branch 'develop' into uif2-initial

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k]
 void add_device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 GNHWC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1p0_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1s1p0_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1s1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1p0_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1s1p0_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1s1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k]
 void add_device_grouped_conv1d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 GNHWC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k]
 void add_device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 GNHWC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k]
 void add_device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 GNHWC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp

@@ -10,7 +10,7 @@ namespace device {
 namespace instance {
 // Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<2,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
                                                                 GKYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_1x1p0_instance.cpp

@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho,
 // wo, k]
 void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_1x1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<3,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 GNDHWC,
                                                                 GKZYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_1x1s1p0_instance.cpp

@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho,
 // wo, k]
 void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_1x1s1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<3,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 GNDHWC,
                                                                 GKZYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_instance.cpp

@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho,
 // wo, k]
 void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<3,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 GNDHWC,
                                                                 GKZYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instance.cpp

@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho,
 // wo, k]
 void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_f16_oddc_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<3,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 GNDHWC,
                                                                 GKZYXC,
                                                                 Empty_Tuple,

library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_1x1p0_instance.cpp

@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for in[g, n, di, hi, wi, c] * wei[g, k, z, y, x, c] = out[g, n, do, ho,
 // wo, k]
 void add_device_grouped_conv3d_fwd_wmma_gndhwc_gkzyxc_gndhwk_i8_1x1p0_instances(
-    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleD<3,
+    std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 GNDHWC,
                                                                 GKZYXC,
                                                                 Empty_Tuple,