diff --git a/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td b/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td
--- a/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td
+++ b/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td
@@ -961,7 +961,8 @@
           "{}">:$transpose_paddings,
          DefaultValuedAttr<StrAttr, "::mlir::bufferization::CopyTensorOp::getOperationName()">:$copy_back_op);
   let results = (outs TransformHandleTypeInterface:$padded,
-                      TransformHandleTypeInterface:$pad);
+                      TransformHandleTypeInterface:$pad,
+                      TransformHandleTypeInterface:$copy);
 
   let assemblyFormat =
     "$target attr-dict `:` functional-type(operands, results)";
diff --git a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
--- a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
+++ b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
@@ -1612,7 +1612,7 @@
 transform::PadOp::apply(transform::TransformRewriter &rewriter,
                         transform::TransformResults &results,
                         transform::TransformState &state) {
-  SmallVector<Operation *> paddedOps, padOps;
+  SmallVector<Operation *> paddedOps, padOps, copyBackOps;
 
   for (Operation *target : state.getPayloadOps(getTarget())) {
     auto linalgTarget = dyn_cast<LinalgOp>(target);
@@ -1707,10 +1707,18 @@
     rewriter.replaceOp(linalgTarget, replacements);
     paddedOps.push_back(paddedOp);
     padOps.append(newPadOps.begin(), newPadOps.end());
+    if (options.copyBackOp != LinalgPaddingOptions::CopyBackOp::None) {
+      for (Value v : replacements) {
+        Operation *copyBackOp = v.getDefiningOp();
+        if (llvm::find(copyBackOps, copyBackOp) == copyBackOps.end())
+          copyBackOps.push_back(copyBackOp);
+      }
+    }
   }
 
   results.set(cast<OpResult>(getPadded()), paddedOps);
   results.set(cast<OpResult>(getPad()), padOps);
+  results.set(cast<OpResult>(getCopy()), copyBackOps);
   return DiagnosedSilenceableFailure::success();
 }
 
diff --git a/mlir/python/mlir/dialects/_structured_transform_ops_ext.py b/mlir/python/mlir/dialects/_structured_transform_ops_ext.py
--- a/mlir/python/mlir/dialects/_structured_transform_ops_ext.py
+++ b/mlir/python/mlir/dialects/_structured_transform_ops_ext.py
@@ -524,6 +524,7 @@
 
         pdl_operation_type = pdl.OperationType.get()
         super().__init__(
+            pdl_operation_type,
             pdl_operation_type,
             pdl_operation_type,
             target,
diff --git a/mlir/test/Dialect/Linalg/pad-to-specific-memory-space.mlir b/mlir/test/Dialect/Linalg/pad-to-specific-memory-space.mlir
--- a/mlir/test/Dialect/Linalg/pad-to-specific-memory-space.mlir
+++ b/mlir/test/Dialect/Linalg/pad-to-specific-memory-space.mlir
@@ -49,11 +49,11 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 1]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
   %buffer, %new_ops = transform.structured.bufferize_to_allocation %pad {memory_space = 3} : !transform.any_op
   %2 = transform.bufferization.one_shot_bufferize %arg1 {bufferize_function_boundaries=true} : (!transform.any_op) -> !transform.any_op
 
@@ -106,11 +106,11 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 1]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
   transform.structured.masked_vectorize %pad vector_sizes [10, 12] : !transform.any_op
   %vector_write = transform.structured.match ops{["vector.transfer_write"]} in %arg1 : (!transform.any_op) -> !transform.any_op
   %mask_op = transform.get_parent_op %vector_write {op_name = "vector.mask"} : (!transform.any_op) -> !transform.any_op
diff --git a/mlir/test/Dialect/Linalg/transform-op-hoist-pad-build-packing-loop-nest.mlir b/mlir/test/Dialect/Linalg/transform-op-hoist-pad-build-packing-loop-nest.mlir
--- a/mlir/test/Dialect/Linalg/transform-op-hoist-pad-build-packing-loop-nest.mlir
+++ b/mlir/test/Dialect/Linalg/transform-op-hoist-pad-build-packing-loop-nest.mlir
@@ -17,10 +17,10 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   // In this case, the pad op is actually empty: we only tile the first dimension
   // and it does not have an impact on the RHS operand.
@@ -49,10 +49,10 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[2]
     : (!transform.any_op) -> !transform.any_op
@@ -87,10 +87,10 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[0]
     : (!transform.any_op) -> !transform.any_op
@@ -125,10 +125,10 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[0]
     : (!transform.any_op) -> !transform.any_op
@@ -161,10 +161,10 @@
 
   %matmul_l1, %loops_l1:2 = transform.structured.tile_to_scf_for %matmul [5, 0, 7] : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[2]
     : (!transform.any_op) -> !transform.any_op
diff --git a/mlir/test/Dialect/Linalg/transform-op-hoist-pad.mlir b/mlir/test/Dialect/Linalg/transform-op-hoist-pad.mlir
--- a/mlir/test/Dialect/Linalg/transform-op-hoist-pad.mlir
+++ b/mlir/test/Dialect/Linalg/transform-op-hoist-pad.mlir
@@ -17,11 +17,11 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     copy_back_op = "none"
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   // In this case, the pad op is actually empty: we only tile the first dimension
   // and it does not have an impact on the RHS operand.
@@ -53,11 +53,11 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     copy_back_op = "none"
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[2]
     : (!transform.any_op) -> !transform.op<"tensor.pad">
@@ -96,11 +96,11 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     copy_back_op = "none"
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[0]
     : (!transform.any_op) -> !transform.any_op
@@ -141,11 +141,11 @@
 
   %matmul_l1, %loops_l1 = transform.structured.tile_to_scf_for %matmul [5] : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     copy_back_op = "none"
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[0]
     : (!transform.any_op) -> !transform.any_op
@@ -185,11 +185,11 @@
 
   %matmul_l1, %loops_l1:2 = transform.structured.tile_to_scf_for %matmul [5, 0, 7] : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
-  %matmul_padded, %0 = transform.structured.pad %matmul_l1 {
+  %matmul_padded, %0, %copy_back = transform.structured.pad %matmul_l1 {
     padding_values=[0.0: f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     copy_back_op = "none"
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 
   %pad = transform.get_producer_of_operand %matmul_padded[2]
     : (!transform.any_op) -> !transform.op<"tensor.pad">
diff --git a/mlir/test/Dialect/Linalg/transform-op-pad.mlir b/mlir/test/Dialect/Linalg/transform-op-pad.mlir
--- a/mlir/test/Dialect/Linalg/transform-op-pad.mlir
+++ b/mlir/test/Dialect/Linalg/transform-op-pad.mlir
@@ -25,6 +25,9 @@
   //      CHECK: %[[T5:.*]] = linalg.matmul
   // CHECK-SAME:              ins(%[[T3]], %[[T4]] : tensor<4x7xf32>, tensor<7x5xf32>)
   // CHECK-SAME:              outs(%[[T2]] : tensor<4x5xf32>)
+
+  //      CHECK: %[[T6:.*]] = tensor.extract_slice %[[T5]]
+  //      CHECK: %[[T7:.*]] = bufferization.copy_tensor %[[T6]], %[[T2]]
   %4 = linalg.matmul ins(%1, %2 : tensor<4x?xf32>, tensor<?x5xf32>) outs(%3 : tensor<4x5xf32>) -> tensor<4x5xf32>
   %5 = tensor.insert_slice %4 into %arg2[%iv0, %iv1] [4, 5] [1, 1] : tensor<4x5xf32> into tensor<24x25xf32>
   func.return %5 : tensor<24x25xf32>
@@ -33,11 +36,13 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 0]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.op<"bufferization.copy_tensor">)
+  // expected-remark @below {{1}}
+  test_print_number_of_associated_payload_ir_ops %copy_back : !transform.op<"bufferization.copy_tensor">
 }
 
 // -----
@@ -65,12 +70,12 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pad_to_multiple_of=[2, 2, 1],
     pack_paddings=[1, 1, 0]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -107,11 +112,11 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 0]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -128,11 +133,11 @@
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
   // expected-error @below {{op expects a padding value of type 'f32', got 0 : i32}}
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0: i32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 0]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -149,11 +154,11 @@
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
   // expected-error @below {{expects a padding that parses to 'f32', got "{foo}"}}
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=["{foo}", 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 0]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -173,11 +178,11 @@
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
   // This error is silenceable and is not reported by this transform
   //   {{transform.structured.pad failed to apply}}
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 0]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -228,11 +233,11 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 1]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -278,9 +283,9 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg1: !transform.any_op):
   %0 = transform.structured.match ops{["linalg.matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
-  %padded, %pad = transform.structured.pad %0 {
+  %padded, %pad, %copy_back = transform.structured.pad %0 {
     padding_values=[0.0 : f32, 0.0 : f32, 0.0 : f32],
     padding_dimensions=[0, 1, 2],
     pack_paddings=[1, 1, 1]
-  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  } : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
diff --git a/mlir/test/Dialect/Linalg/transform-ops-invalid.mlir b/mlir/test/Dialect/Linalg/transform-ops-invalid.mlir
--- a/mlir/test/Dialect/Linalg/transform-ops-invalid.mlir
+++ b/mlir/test/Dialect/Linalg/transform-ops-invalid.mlir
@@ -11,7 +11,7 @@
 transform.sequence failures(propagate) {
 ^bb0(%arg0: !transform.any_op):
   // expected-error@below {{expects padding_dimensions to contain positive integers, found [1, -7]}}
-  transform.structured.pad %arg0 {padding_dimensions=[1, -7]} : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  transform.structured.pad %arg0 {padding_dimensions=[1, -7]} : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -19,7 +19,7 @@
 transform.sequence failures(propagate) {
 ^bb0(%arg0: !transform.any_op):
   // expected-error@below {{expects pack_paddings to contain booleans (0/1), found [1, 7]}}
-  transform.structured.pad %arg0 {pack_paddings=[1, 7]} : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  transform.structured.pad %arg0 {pack_paddings=[1, 7]} : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
@@ -27,7 +27,7 @@
 transform.sequence failures(propagate) {
 ^bb0(%arg0: !transform.any_op):
   // expected-error@below {{expects transpose_paddings to be a permutation, found [1, 1]}}
-  transform.structured.pad %arg0 {transpose_paddings=[[1, 1]]} : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  transform.structured.pad %arg0 {transpose_paddings=[[1, 1]]} : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 // -----
diff --git a/mlir/test/Dialect/Linalg/transform-ops.mlir b/mlir/test/Dialect/Linalg/transform-ops.mlir
--- a/mlir/test/Dialect/Linalg/transform-ops.mlir
+++ b/mlir/test/Dialect/Linalg/transform-ops.mlir
@@ -21,7 +21,7 @@
 transform.sequence failures(propagate) {
 ^bb1(%arg0: !transform.any_op):
   // CHECK: transform.structured.pad
-  %0, %1 = transform.structured.pad %arg0 : (!transform.any_op) -> (!transform.any_op, !transform.any_op)
+  %0, %1, %2 = transform.structured.pad %arg0 : (!transform.any_op) -> (!transform.any_op, !transform.any_op, !transform.any_op)
 }
 
 transform.sequence failures(propagate) {