diff --git a/tensorflow/core/kernels/mkl/mkl_batch_matmul_op.cc b/tensorflow/core/kernels/mkl/mkl_batch_matmul_op.cc
index 6b526264496e80..05ff773271cc9e 100644
--- a/tensorflow/core/kernels/mkl/mkl_batch_matmul_op.cc
+++ b/tensorflow/core/kernels/mkl/mkl_batch_matmul_op.cc
@@ -38,7 +38,6 @@ limitations under the License.
 #include "tensorflow/core/platform/errors.h"
 #include "tensorflow/core/platform/types.h"
 #include "tensorflow/core/util/matmul_bcast.h"
-#include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor"
 
 namespace tensorflow {
 
diff --git a/tensorflow/core/kernels/mkl/mkl_matmul_ops_common.h b/tensorflow/core/kernels/mkl/mkl_matmul_ops_common.h
index 6a652ba0ec5273..5f74e09e97d7d5 100644
--- a/tensorflow/core/kernels/mkl/mkl_matmul_ops_common.h
+++ b/tensorflow/core/kernels/mkl/mkl_matmul_ops_common.h
@@ -21,7 +21,6 @@ limitations under the License.
 #include <string>
 #include <vector>
 
-#include "unsupported/Eigen/CXX11/Tensor"  // from @eigen_archive
 #include "dnnl.hpp"
 #include "tensorflow/core/framework/op.h"
 #include "tensorflow/core/framework/op_kernel.h"
@@ -29,6 +28,7 @@ limitations under the License.
 #include "tensorflow/core/kernels/mkl/mkl_kernel_util.h"
 #include "tensorflow/core/util/mkl_util.h"
 #include "tensorflow/core/util/onednn_env_vars.h"
+#include "unsupported/Eigen/CXX11/Tensor"  // from @eigen_archive
 #if defined(DNNL_AARCH64_USE_ACL) && defined(ENABLE_ONEDNN_OPENMP)
 #include "tensorflow/core/platform/mutex.h"
 #endif  // DNNL_AARCH64_USE_ACL && ENABLE_ONEDNN_OPENMP
@@ -814,7 +814,7 @@ class MklMatMulPrimitive : public MklPrimitive {
   void Execute(const std::shared_ptr<stream>& stream, const Tlhs* a_data,
                const Trhs* b_data, const Toutput* c_data,
                const MklMatMulParams& matmul_params, void* sp_data,
-               const std::vector<void*> binary_op_fusions_data = {}) {
+               const std::vector<void*>& binary_op_fusions_data = {}) {
 #if defined(DNNL_AARCH64_USE_ACL) && defined(ENABLE_ONEDNN_OPENMP)
     mutex_lock lock(primitive_execution_mu_);
 #endif
@@ -829,9 +829,10 @@ class MklMatMulPrimitive : public MklPrimitive {
         static_cast<void*>(const_cast<Toutput*>(c_data)), *stream);
     context_.sp_mem->set_data_handle(sp_data, *stream);
 
-    for (int i = 0; i < num_post_ops_data; ++i)
+    for (int i = 0; i < num_post_ops_data; ++i) {
       context_.post_ops_mem[i]->set_data_handle(binary_op_fusions_data[i],
                                                 *stream);
+    }
 #else
     context_.a_mem->set_data_handle(
         static_cast<void*>(const_cast<Tlhs*>(a_data)));
@@ -861,8 +862,9 @@ class MklMatMulPrimitive : public MklPrimitive {
     context_.b_mem->set_data_handle(DummyData);
     context_.c_mem->set_data_handle(DummyData);
     context_.sp_mem->set_data_handle(DummyData);
-    for (int i = 0; i < num_post_ops_data; ++i)
+    for (int i = 0; i < num_post_ops_data; ++i) {
       context_.post_ops_mem[i]->set_data_handle(DummyData);
+    }
   }
 
   std::shared_ptr<dnnl::matmul::primitive_desc> GetPrimitiveDesc() const {
@@ -878,7 +880,7 @@ class MklMatMulPrimitive : public MklPrimitive {
     std::shared_ptr<dnnl::memory> c_mem;
     std::shared_ptr<dnnl::memory> sp_mem;
 
-    // Quantization scale related memory
+    // Quantization scale related memory.
     std::shared_ptr<dnnl::memory> lhs_scale_mem;
     std::shared_ptr<dnnl::memory> rhs_scale_mem;
     std::shared_ptr<dnnl::memory> dst_scale_mem;
@@ -896,7 +898,7 @@ class MklMatMulPrimitive : public MklPrimitive {
     std::shared_ptr<dnnl::memory::desc> b_md;
     std::shared_ptr<dnnl::memory::desc> c_md;
 
-    // Quantization scale related memory descriptors
+    // Quantization scale related memory descriptors.
     std::shared_ptr<dnnl::memory::desc> lhs_scale_md;
     std::shared_ptr<dnnl::memory::desc> rhs_scale_md;
     std::shared_ptr<dnnl::memory::desc> dst_scale_md;
@@ -969,7 +971,7 @@ class MklMatMulPrimitive : public MklPrimitive {
     dnnl::primitive_attr post_ops_attr;
     dnnl::post_ops post_ops;
     std::unordered_map<string, bool> is_scale_set;
-    int binary_post_ops_count = 0;  // Keep track op binary fusions
+    int binary_post_ops_count = 0;  // Keep track op binary fusions.
     if (!post_op_params.empty()) {
       for (auto const& post_op_param : post_op_params) {
         if (post_op_param.name == "lhs_scale") {