diff --git a/flang/lib/Optimizer/Transforms/CUFOpConversion.cpp b/flang/lib/Optimizer/Transforms/CUFOpConversion.cpp
index b14a1c338e087f..c070c3de94cf81 100644
--- a/flang/lib/Optimizer/Transforms/CUFOpConversion.cpp
+++ b/flang/lib/Optimizer/Transforms/CUFOpConversion.cpp
@@ -640,15 +640,20 @@ struct CUFDataTransferOpConversion
                     loc, builder);
       mlir::Value dst = op.getDst();
       mlir::Value src = op.getSrc();
-
       if (!mlir::isa<fir::BaseBoxType>(srcTy)) {
         src = emboxSrc(rewriter, op, symtab);
-      } else if (mlir::isa<fir::EmboxOp>(src.getDefiningOp())) {
-        // Materialize the box to memory to be able to call the runtime.
-        mlir::Value box = builder.createTemporary(loc, src.getType());
-        builder.create<fir::StoreOp>(loc, src, box);
-        src = box;
       }
+      auto materializeBoxIfNeeded = [&](mlir::Value val) -> mlir::Value {
+        if (mlir::isa<fir::EmboxOp>(val.getDefiningOp())) {
+          // Materialize the box to memory to be able to call the runtime.
+          mlir::Value box = builder.createTemporary(loc, val.getType());
+          builder.create<fir::StoreOp>(loc, val, box);
+          return box;
+        }
+        return val;
+      };
+      src = materializeBoxIfNeeded(src);
+      dst = materializeBoxIfNeeded(dst);
 
       auto fTy = func.getFunctionType();
       mlir::Value sourceFile = fir::factory::locationToFilename(builder, loc);
diff --git a/flang/test/Fir/CUDA/cuda-data-transfer.fir b/flang/test/Fir/CUDA/cuda-data-transfer.fir
index 69baf7d15a7d03..ad392beed56d47 100644
--- a/flang/test/Fir/CUDA/cuda-data-transfer.fir
+++ b/flang/test/Fir/CUDA/cuda-data-transfer.fir
@@ -400,7 +400,6 @@ func.func @_QPdevmul(%arg0: !fir.ref<!fir.array<1x?xf32>> {fir.bindc_name = "b"}
   %9 = fir.convert %8 : (i32) -> index
   %12 = fir.shape %c1, %9 : (index, index) -> !fir.shape<2>
   %13 = fir.declare %arg0(%12) dummy_scope %0 {uniq_name = "_QFdevmulEb"} : (!fir.ref<!fir.array<1x?xf32>>, !fir.shape<2>, !fir.dscope) -> !fir.ref<!fir.array<1x?xf32>>
-
   %24 = fir.load %7 : !fir.ref<i32>
   %25 = fir.convert %24 : (i32) -> index
   %26 = arith.cmpi sgt, %25, %c0 : index
@@ -414,14 +413,19 @@ func.func @_QPdevmul(%arg0: !fir.ref<!fir.array<1x?xf32>> {fir.bindc_name = "b"}
   %34 = fir.slice %c1, %25, %c1, %c1, %29, %c1 : (index, index, index, index, index, index) -> !fir.slice<2>
   %35 = fir.embox %13(%12) [%34] : (!fir.ref<!fir.array<1x?xf32>>, !fir.shape<2>, !fir.slice<2>) -> !fir.box<!fir.array<?x?xf32>>
   cuf.data_transfer %35 to %6 {transfer_kind = #cuf.cuda_transfer<host_device>} : !fir.box<!fir.array<?x?xf32>>, !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xf32>>>>
+  cuf.data_transfer %6 to %35 {transfer_kind = #cuf.cuda_transfer<device_host>} : !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xf32>>>>, !fir.box<!fir.array<?x?xf32>>
   return
 }
 
 // CHECK-LABEL: func.func @_QPdevmul(%arg0: !fir.ref<!fir.array<1x?xf32>> {fir.bindc_name = "b"}, %arg1: !fir.ref<i32> {fir.bindc_name = "wa"}, %arg2: !fir.ref<i32> {fir.bindc_name = "wb"}) {
-// CHECK: %[[ALLOCA:.*]] = fir.alloca !fir.box<!fir.array<?x?xf32>>
+// CHECK: %[[ALLOCA0:.*]] = fir.alloca !fir.box<!fir.array<?x?xf32>>
+// CHECK: %[[ALLOCA1:.*]] = fir.alloca !fir.box<!fir.array<?x?xf32>>
 // CHECK: %[[EMBOX:.*]] = fir.embox %{{.*}}(%{{.*}}) [%{{.*}}] : (!fir.ref<!fir.array<1x?xf32>>, !fir.shape<2>, !fir.slice<2>) -> !fir.box<!fir.array<?x?xf32>>
-// CHECK: fir.store %[[EMBOX]] to %[[ALLOCA]] : !fir.ref<!fir.box<!fir.array<?x?xf32>>>
-// CHECK: %[[SRC:.*]] = fir.convert %[[ALLOCA]] : (!fir.ref<!fir.box<!fir.array<?x?xf32>>>) -> !fir.ref<!fir.box<none>>
+// CHECK: fir.store %[[EMBOX]] to %[[ALLOCA1]] : !fir.ref<!fir.box<!fir.array<?x?xf32>>>
+// CHECK: %[[SRC:.*]] = fir.convert %[[ALLOCA1]] : (!fir.ref<!fir.box<!fir.array<?x?xf32>>>) -> !fir.ref<!fir.box<none>>
 // CHECK: fir.call @_FortranACUFDataTransferDescDesc(%{{.*}}, %[[SRC]], %{{.*}}, %{{.*}}, %{{.*}}) : (!fir.ref<!fir.box<none>>, !fir.ref<!fir.box<none>>, i32, !fir.ref<i8>, i32) -> none
+// CHECK: fir.store %[[EMBOX]] to %[[ALLOCA0]] : !fir.ref<!fir.box<!fir.array<?x?xf32>>>
+// CHECK: %[[DST:.*]] = fir.convert %[[ALLOCA0]] : (!fir.ref<!fir.box<!fir.array<?x?xf32>>>) -> !fir.ref<!fir.box<none>>
+// CHECK: fir.call @_FortranACUFDataTransferDescDesc(%[[DST]], %{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}) : (!fir.ref<!fir.box<none>>, !fir.ref<!fir.box<none>>, i32, !fir.ref<i8>, i32) -> none
 
 } // end of module