diff --git a/python/paddle/fluid/tests/unittests/mkldnn/test_prelu_mkldnn_op.py b/python/paddle/fluid/tests/unittests/mkldnn/test_prelu_mkldnn_op.py
index 5489bf109dd54aea3440e66811f75960ed117fc7..901aa200a377511bcd7e69cc4309e813a1ff8a47 100644
--- a/python/paddle/fluid/tests/unittests/mkldnn/test_prelu_mkldnn_op.py
+++ b/python/paddle/fluid/tests/unittests/mkldnn/test_prelu_mkldnn_op.py
@@ -18,7 +18,7 @@ import unittest
 import numpy as np
 import paddle
 import paddle.fluid.core as core
-from paddle.fluid.tests.unittests.op_test import OpTest, convert_float_to_uint16
+from paddle.fluid.tests.unittests.op_test import OpTest, OpTestTool, convert_float_to_uint16
 
 
 def ref_prelu(x, weight, mode):
@@ -109,6 +109,7 @@ class TestPReluModeAllAlpha1DOneDNNOp(TestPReluModeAllOneDNNOp):
 
 #   BF16 TESTS
 def create_bf16_test_class(parent):
+    @OpTestTool.skip_if_not_cpu_bf16()
     class TestPReluBF16OneDNNOp(parent):
         def set_inputs(self, ):
             self.inputs = {
@@ -143,42 +144,25 @@ def create_bf16_test_class(parent):
             self.dout = dout
 
         def test_check_output(self):
-            if core.is_compiled_with_cuda():
-                self.skipTest(
-                    "OneDNN doesn't support bf16 with CUDA, skipping UT" +
-                    self.__class__.__name__)
-            elif not core.supports_bfloat16():
-                self.skipTest("Core doesn't support bf16, skipping UT" +
-                              self.__class__.__name__)
-            else:
-                self.check_output_with_place(core.CPUPlace())
+            self.check_output_with_place(core.CPUPlace())
 
         def test_check_grad(self):
-            if core.is_compiled_with_cuda() or not core.supports_bfloat16():
-                self.skipTest(
-                    "Core is compiled with cuda or doesn't support bf16, kipping UT"
-                    + self.__class__.__name__)
-            else:
-                self.calculate_grads()
-                self.check_grad_with_place(
-                    core.CPUPlace(), ["X", "Alpha"],
-                    "Out",
-                    user_defined_grads=[self.dx, self.dalpha],
-                    user_defined_grad_outputs=[
-                        convert_float_to_uint16(self.dout)
-                    ])
+            self.calculate_grads()
+            self.check_grad_with_place(
+                core.CPUPlace(), ["X", "Alpha"],
+                "Out",
+                user_defined_grads=[self.dx, self.dalpha],
+                user_defined_grad_outputs=[convert_float_to_uint16(self.dout)])
 
     cls_name = "{0}_{1}".format(parent.__name__, "BF16")
     TestPReluBF16OneDNNOp.__name__ = cls_name
     globals()[cls_name] = TestPReluBF16OneDNNOp
 
 
-#TODO jakpiase
-#enable bf16 tests back when oneDNN bf16 class will be ready
-#create_bf16_test_class(TestPReluModeChannelOneDNNOp)
-#create_bf16_test_class(TestPReluModeElementOneDNNOp)
-#create_bf16_test_class(TestPReluModeChannel3DOneDNNOp)
-#create_bf16_test_class(TestPReluModeChannelAlpha1DOneDNNOp)
+create_bf16_test_class(TestPReluModeChannelOneDNNOp)
+create_bf16_test_class(TestPReluModeElementOneDNNOp)
+create_bf16_test_class(TestPReluModeChannel3DOneDNNOp)
+create_bf16_test_class(TestPReluModeChannelAlpha1DOneDNNOp)
 
 if __name__ == "__main__":
     paddle.enable_static()