Update the 'missing instructions' message. (Stock TF calls MKL sgemm for matmul-like operations now).

PiperOrigin-RevId: 301250082
Change-Id: I2af16c8019e21a0c42e6b6f0336471f4b3139a3c
diff --git a/tensorflow/core/platform/cpu_feature_guard.cc b/tensorflow/core/platform/cpu_feature_guard.cc
index 235dc57..c5a5c28 100644
--- a/tensorflow/core/platform/cpu_feature_guard.cc
+++ b/tensorflow/core/platform/cpu_feature_guard.cc
@@ -139,16 +139,11 @@
 #endif  // __FMA__
 #endif  // else of if defined(_MSC_VER) && !defined(__clang__)
     if (!missing_instructions.empty()) {
-#ifndef INTEL_MKL
-      LOG(INFO) << "Your CPU supports instructions that this TensorFlow "
-                << "binary was not compiled to use:" << missing_instructions;
-#else
       LOG(INFO) << "This TensorFlow binary is optimized with Intel(R) MKL-DNN "
-                << "to use the following CPU instructions in performance "
+                << "to use the following CPU instructions in performance-"
                 << "critical operations: " << missing_instructions << std::endl
-                << "To enable them in non-MKL-DNN operations, rebuild "
-                << "TensorFlow with the appropriate compiler flags.";
-#endif
+                << "To enable them in other operations, rebuild TensorFlow "
+                << "with the appropriate compiler flags.";
     }
   });
 }