diff --git a/kernels/portable/cpu/op_detach_copy.cpp b/kernels/portable/cpu/op_detach_copy.cpp
index 844f259f6de..a8db4b0804b 100644
--- a/kernels/portable/cpu/op_detach_copy.cpp
+++ b/kernels/portable/cpu/op_detach_copy.cpp
@@ -33,6 +33,9 @@ Tensor& detach_copy_out(RuntimeContext& ctx, const Tensor& self, Tensor& out) {
       out,
       "Failed to resize output tensor.");
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(self, out), InvalidArgument, out);
+
   ET_KERNEL_CHECK(
       ctx, tensors_have_same_shape_and_dtype(self, out), InvalidArgument, out);
 
diff --git a/kernels/portable/cpu/op_diagonal_copy.cpp b/kernels/portable/cpu/op_diagonal_copy.cpp
index 67b14c3f792..0de86ea0a64 100644
--- a/kernels/portable/cpu/op_diagonal_copy.cpp
+++ b/kernels/portable/cpu/op_diagonal_copy.cpp
@@ -73,6 +73,11 @@ Tensor& diagonal_copy_out(
   ET_KERNEL_CHECK(
       ctx, check_diagonal_copy_args(in, dim1, dim2, out), InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(in, out), InvalidArgument, out);
+
+  ET_KERNEL_CHECK(ctx, tensor_is_default_dim_order(in), InvalidArgument, out);
+
   if (dim1 < 0) {
     dim1 += nonzero_dim(in);
   }
diff --git a/kernels/portable/cpu/op_div.cpp b/kernels/portable/cpu/op_div.cpp
index 84591cb0ebd..db2079c2ff2 100644
--- a/kernels/portable/cpu/op_div.cpp
+++ b/kernels/portable/cpu/op_div.cpp
@@ -41,6 +41,9 @@ div_out(RuntimeContext& ctx, const Tensor& a, const Tensor& b, Tensor& out) {
       InvalidArgument,
       out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, b, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
 
@@ -97,6 +100,9 @@ Tensor& div_out_mode(
       InvalidArgument,
       out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, b, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
   ScalarType common_type = get_compute_type(a_type, b_type);
@@ -159,6 +165,9 @@ Tensor& div_scalar_out(
   ScalarType common_type = isFloatingType(a_type) ? a_type : ScalarType::Float;
   ScalarType out_type = out.scalar_type();
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ET_KERNEL_CHECK(ctx, common_type == out_type, InvalidArgument, out);
 
   ET_SWITCH_REAL_TYPES_AND(Bool, a_type, ctx, "div.Scalar_out", CTYPE_A, [&]() {
diff --git a/kernels/portable/cpu/op_embedding.cpp b/kernels/portable/cpu/op_embedding.cpp
index ffa43da7395..1b493435af5 100644
--- a/kernels/portable/cpu/op_embedding.cpp
+++ b/kernels/portable/cpu/op_embedding.cpp
@@ -102,6 +102,15 @@ Tensor& embedding_out(
       out.size(1),
       weight.size(1));
 
+  ET_KERNEL_CHECK(
+      ctx,
+      tensors_have_same_dim_order(weight, indices, out),
+      InvalidArgument,
+      out);
+
+  ET_KERNEL_CHECK(
+      ctx, tensor_is_default_dim_order(weight), InvalidArgument, out);
+
   ScalarType ix_type = indices.scalar_type();
   ET_CHECK_MSG(
       ix_type == ScalarType::Long || ix_type == ScalarType::Int,
diff --git a/kernels/portable/cpu/op_eq.cpp b/kernels/portable/cpu/op_eq.cpp
index 8a4e4656f08..8fef70e564f 100644
--- a/kernels/portable/cpu/op_eq.cpp
+++ b/kernels/portable/cpu/op_eq.cpp
@@ -34,6 +34,9 @@ Tensor& eq_tensor_out(
   ScalarType b_type = b.scalar_type();
   ScalarType out_type = out.scalar_type();
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, b, out), InvalidArgument, out);
+
   ET_SWITCH_REAL_TYPES_AND(Bool, a_type, ctx, "eq.Scalar_out", CTYPE_A, [&]() {
     ET_SWITCH_REAL_TYPES_AND(
         Bool, b_type, ctx, "eq.Scalar_out", CTYPE_B, [&]() {
@@ -80,6 +83,9 @@ Tensor& eq_scalar_out(
   ScalarType b_type = utils::get_scalar_dtype(b);
   ScalarType out_type = out.scalar_type();
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ET_SWITCH_REAL_TYPES_AND(Bool, a_type, ctx, "eq.Scalar_out", CTYPE_A, [&]() {
     ET_SWITCH_SCALAR_OBJ_TYPES(b_type, ctx, "eq.Scalar_out", CTYPE_B, [&]() {
       using CTYPE_IN =
diff --git a/kernels/portable/cpu/op_expand_copy.cpp b/kernels/portable/cpu/op_expand_copy.cpp
index 5f0d19adc59..67e9149ff9f 100644
--- a/kernels/portable/cpu/op_expand_copy.cpp
+++ b/kernels/portable/cpu/op_expand_copy.cpp
@@ -85,6 +85,10 @@ Tensor& expand_copy_out(
       InvalidArgument,
       out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(self, out), InvalidArgument, out);
+  ET_KERNEL_CHECK(ctx, tensor_is_default_dim_order(self), InvalidArgument, out);
+
   // Holds the result of expand_sizes converted to repeat sizes
   int64_t repeats[kTensorDimensionLimit];
   const auto repeats_size{map_expand_to_repeats(
diff --git a/kernels/portable/cpu/op_fill.cpp b/kernels/portable/cpu/op_fill.cpp
index 60ebd5de5ab..d908c53e0c7 100644
--- a/kernels/portable/cpu/op_fill.cpp
+++ b/kernels/portable/cpu/op_fill.cpp
@@ -31,6 +31,9 @@ Tensor& fill_scalar_out(
 
   ET_KERNEL_CHECK(ctx, a_type == out_type, InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   // Resize for dynamic shape
   ET_KERNEL_CHECK_MSG(
       ctx,
@@ -67,6 +70,9 @@ Tensor& fill_tensor_out(
   // Assert `b` must be a scalar tensor.
   ET_KERNEL_CHECK(ctx, tensor_is_scalar(b), InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
   ScalarType out_type = out.scalar_type();
diff --git a/kernels/portable/cpu/op_flip.cpp b/kernels/portable/cpu/op_flip.cpp
index 10c52439d11..c88585f88a5 100644
--- a/kernels/portable/cpu/op_flip.cpp
+++ b/kernels/portable/cpu/op_flip.cpp
@@ -45,6 +45,9 @@ flip_out(RuntimeContext& ctx, const Tensor& in, IntArrayRef dims, Tensor& out) {
   ET_KERNEL_CHECK(
       ctx, resize_tensor(out, in.sizes()) == Error::Ok, InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(in, out), InvalidArgument, out);
+
   ET_KERNEL_CHECK(ctx, check_flip_args(in, dims, out), InvalidArgument, out);
 
   bool flip_dim_data[kTensorDimensionLimit];
diff --git a/kernels/portable/cpu/op_floor_divide.cpp b/kernels/portable/cpu/op_floor_divide.cpp
index 0514df0ca25..88c6d5e7e79 100644
--- a/kernels/portable/cpu/op_floor_divide.cpp
+++ b/kernels/portable/cpu/op_floor_divide.cpp
@@ -87,6 +87,9 @@ Tensor& floor_divide_out(
 
   ET_KERNEL_CHECK(ctx, tensor_is_real_type(out), InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, b, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
   ScalarType common_type = promoteTypes(a_type, b_type);
diff --git a/kernels/portable/cpu/op_fmod.cpp b/kernels/portable/cpu/op_fmod.cpp
index 42f83731199..6743eb8cf8a 100644
--- a/kernels/portable/cpu/op_fmod.cpp
+++ b/kernels/portable/cpu/op_fmod.cpp
@@ -85,6 +85,9 @@ Tensor& fmod_Tensor_out(
       InvalidArgument,
       out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, b, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
   ScalarType common_type = promoteTypes(a_type, b_type);
@@ -139,6 +142,9 @@ Tensor& fmod_Scalar_out(
       out,
       "Failed to resize output tensor.");
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = utils::get_scalar_dtype(b);
   ScalarType common_type = utils::promote_type_with_scalar(a_type, b);
diff --git a/kernels/portable/cpu/op_full_like.cpp b/kernels/portable/cpu/op_full_like.cpp
index 880e02efe66..0ce8923ccdf 100644
--- a/kernels/portable/cpu/op_full_like.cpp
+++ b/kernels/portable/cpu/op_full_like.cpp
@@ -34,6 +34,11 @@ Tensor& full_like_out(
         "memory_format must be contiguous");
   }
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(in, out), InvalidArgument, out);
+
+  ET_KERNEL_CHECK(ctx, tensor_is_default_dim_order(in), InvalidArgument, out);
+
   // Resize for dynamic shape
   ET_KERNEL_CHECK_MSG(
       ctx,
diff --git a/kernels/portable/cpu/op_ge.cpp b/kernels/portable/cpu/op_ge.cpp
index 88e056e7362..d89c45cca45 100644
--- a/kernels/portable/cpu/op_ge.cpp
+++ b/kernels/portable/cpu/op_ge.cpp
@@ -31,6 +31,9 @@ Tensor& ge_tensor_out(
       InvalidArgument,
       out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
   ScalarType out_type = out.scalar_type();
@@ -77,6 +80,9 @@ Tensor& ge_scalar_out(
       out,
       "Failed to resize output tensor.");
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = utils::get_scalar_dtype(b);
   ScalarType common_type = utils::promote_type_with_scalar(a_type, b);
diff --git a/kernels/portable/cpu/op_gelu.cpp b/kernels/portable/cpu/op_gelu.cpp
index 0432c028141..4fadd2aff58 100644
--- a/kernels/portable/cpu/op_gelu.cpp
+++ b/kernels/portable/cpu/op_gelu.cpp
@@ -34,6 +34,9 @@ Tensor& gelu_out(
   ET_KERNEL_CHECK(
       ctx, resize_tensor(out, in.sizes()) == Error::Ok, InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(in, out), InvalidArgument, out);
+
   ET_SWITCH_FLOAT_TYPES(in.scalar_type(), ctx, "gelu.out", CTYPE, [&]() {
     if (approximate == "tanh") {
       apply_unary_map_fn(
diff --git a/kernels/portable/cpu/op_glu.cpp b/kernels/portable/cpu/op_glu.cpp
index 5a075ff35ca..4ecdbbc1caf 100644
--- a/kernels/portable/cpu/op_glu.cpp
+++ b/kernels/portable/cpu/op_glu.cpp
@@ -144,6 +144,9 @@ glu_out(RuntimeContext& ctx, const Tensor& self, int64_t dim, Tensor& out) {
   ET_KERNEL_CHECK(
       ctx, resize_glu_out(self, dim, out) == Error::Ok, InvalidArgument, out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(self, out), InvalidArgument, out);
+
   ET_KERNEL_CHECK(ctx, check_glu_args(self, dim, out), InvalidArgument, out);
 
   const size_t non_negative_dim = dim < 0 ? dim + self.dim() : dim;
diff --git a/kernels/portable/cpu/op_gt.cpp b/kernels/portable/cpu/op_gt.cpp
index 56d8657c9b5..4c5df64cb69 100644
--- a/kernels/portable/cpu/op_gt.cpp
+++ b/kernels/portable/cpu/op_gt.cpp
@@ -31,6 +31,9 @@ Tensor& gt_tensor_out(
       InvalidArgument,
       out);
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, b, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = b.scalar_type();
   ScalarType out_type = out.scalar_type();
@@ -77,6 +80,9 @@ Tensor& gt_scalar_out(
       out,
       "Failed to resize output tensor.");
 
+  ET_KERNEL_CHECK(
+      ctx, tensors_have_same_dim_order(a, out), InvalidArgument, out);
+
   ScalarType a_type = a.scalar_type();
   ScalarType b_type = utils::get_scalar_dtype(b);
   ScalarType common_type = utils::promote_type_with_scalar(a_type, b);