Fix security vulnerability with FractionalMax(AVG)Pool with illegal pooling_ratio

PiperOrigin-RevId: 483486453

tensorflow/core/kernels/fractional_avg_pool_op.cc

@@ -44,6 +44,12 @@ class FractionalAvgPoolOp : public OpKernel {
     OP_REQUIRES(context, pooling_ratio_.size() == 4,
                 errors::InvalidArgument(
                     "pooling_ratio field must specify 4 dimensions"));
+    for (std::size_t i = 0; i < pooling_ratio_.size(); ++i) {
+      OP_REQUIRES(context, pooling_ratio_[i] >= 1,
+                  errors::InvalidArgument(
+                      "pooling_ratio cannot be smaller than 1, got: ",
+                      pooling_ratio_[i]));
+    }
     OP_REQUIRES(
         context, pooling_ratio_[0] == 1 || pooling_ratio_[3] == 1,
         errors::Unimplemented("Fractional average pooling is not yet "
@@ -82,9 +88,11 @@ class FractionalAvgPoolOp : public OpKernel {
     for (int i = 0; i < tensor_in_and_out_dims; ++i) {
       input_size[i] = tensor_in.dim_size(i);
       OP_REQUIRES(
-          context, pooling_ratio_[i] <= input_size[i],
-          errors::InvalidArgument(
-              "Pooling ratio cannot be bigger than input tensor dim size."));
+          context, input_size[i] >= pooling_ratio_[i],
+          errors::InvalidArgument("Pooling ratio is higher than input "
+                                  "dimension size for dimension ",
+                                  i, ". Input dim size: ", input_size[i],
+                                  " pooling ratio: ", pooling_ratio_[i]));
     }
     // Output size.
     for (int i = 0; i < tensor_in_and_out_dims; ++i) {

tensorflow/core/kernels/fractional_max_pool_op.cc

@@ -45,6 +45,12 @@ class FractionalMaxPoolOp : public OpKernel {
     OP_REQUIRES(context, pooling_ratio_.size() == 4,
                 errors::InvalidArgument("pooling_ratio field must "
                                         "specify 4 dimensions"));
+    for (std::size_t i = 0; i < pooling_ratio_.size(); ++i) {
+      OP_REQUIRES(context, pooling_ratio_[i] >= 1,
+                  errors::InvalidArgument(
+                      "pooling_ratio cannot be smaller than 1, got: ",
+                      pooling_ratio_[i]));
+    }
 
     OP_REQUIRES(
         context, pooling_ratio_[0] == 1 || pooling_ratio_[3] == 1,

tensorflow/core/ops/nn_ops.cc

@@ -63,6 +63,13 @@ Status FractionalPoolShapeFn(InferenceContext* c) {
     }
   }
 
+  for (std::size_t i = 0; i < pooling_ratio.size(); ++i) {
+    if (pooling_ratio[i] < 1) {
+      return errors::InvalidArgument(
+          "pooling_ratio cannot be smaller than 1, got: ", pooling_ratio[i]);
+    }
+  }
+
   c->set_output(0, c->MakeShape(output_dims));
   c->set_output(1, c->Vector(output_dims[1]));
   c->set_output(2, c->Vector(output_dims[2]));

tensorflow/core/ops/nn_ops_test.cc

@@ -523,7 +523,8 @@ TEST(NNOpsTest, FractionalPool_ShapeFn) {
                        .Finalize(&op.node_def));
     };
 
-    set_op(std::vector<float>{2.0f, 1, 1 / 1.5f, 1 / 2.0f});
+    // pooling_ratio must >= 1.0
+    set_op(std::vector<float>{2.0f, 1, 1.5f, 4.0f});
 
     // Rank check.
     INFER_ERROR("must be rank 4", op, "[?,?,?]");
@@ -532,11 +533,11 @@ TEST(NNOpsTest, FractionalPool_ShapeFn) {
     INFER_OK(op, "?", "[?,?,?,?];[?];[?]");
     INFER_OK(op, "[?,?,?,?]", "[?,?,?,?];[?];[?]");
 
-    INFER_OK(op, "[10,20,30,40]", "[5,20,45,80];[20];[45]");
-    INFER_OK(op, "[?,20,30,40]", "[?,20,45,80];[20];[45]");
-    INFER_OK(op, "[10,?,30,40]", "[5,?,45,80];[?];[45]");
-    INFER_OK(op, "[10,20,?,40]", "[5,20,?,80];[20];[?]");
-    INFER_OK(op, "[10,20,30,?]", "[5,20,45,?];[20];[45]");
+    INFER_OK(op, "[10,20,30,40]", "[5,20,20,10];[20];[20]");
+    INFER_OK(op, "[?,20,30,40]", "[?,20,20,10];[20];[20]");
+    INFER_OK(op, "[10,?,30,40]", "[5,?,20,10];[?];[20]");
+    INFER_OK(op, "[10,20,?,40]", "[5,20,?,10];[20];[?]");
+    INFER_OK(op, "[10,20,30,?]", "[5,20,20,?];[20];[20]");
 
     // Wrong number of values for pooling_ratio.
     set_op(std::vector<float>{.5, 1.0, 1.5});

tensorflow/python/kernel_tests/nn_ops/fractional_avg_pool_op_test.py

@@ -333,6 +333,41 @@ class FractionalAvgTest(test.TestCase):
 
         self.evaluate(z)
 
+  def testPoolingRatioHasMoreDimThanInput(self):
+    with self.cached_session() as _:
+      with self.assertRaisesRegex(
+          errors.InvalidArgumentError,
+          r"Pooling ratio is higher than input dimension size for dimension 1.*"
+      ):
+        result = nn_ops.gen_nn_ops.fractional_avg_pool(
+            value=constant_op.constant(
+                value=[[[[1, 4, 2, 3]]]], dtype=dtypes.int64),
+            pooling_ratio=[1.0, 1.44, 1.73, 1.0],
+            pseudo_random=False,
+            overlapping=False,
+            deterministic=False,
+            seed=0,
+            seed2=0,
+            name=None)
+        self.evaluate(result)
+
+  def testPoolingRatioValueOutOfRange(self):
+    with self.cached_session() as _:
+      # Whether turn on `TF2_BEHAVIOR` generates different error messages
+      with self.assertRaisesRegex(
+          (errors.InvalidArgumentError, ValueError),
+          r"(pooling_ratio cannot be smaller than 1, got: .*)|(is negative)"):
+        result = nn_ops.gen_nn_ops.fractional_avg_pool(
+            value=np.zeros([3, 30, 30, 3]),
+            pooling_ratio=[1, -1, 3, 1],
+            pseudo_random=False,
+            overlapping=False,
+            deterministic=False,
+            seed=0,
+            seed2=0,
+        )
+        self.evaluate(result)
+
 
 class FractionalAvgPoolGradTest(test.TestCase):
   """Tests for FractionalAvgPoolGrad.

tensorflow/python/kernel_tests/nn_ops/fractional_max_pool_op_test.py

@@ -320,7 +320,7 @@ class FractionalMaxPoolTest(test.TestCase):
       nn_ops.fractional_max_pool(
           rand_mat, [1, 1.5, 1.5, 1], seed=1, seed2=1, deterministic=True)
 
-  def testPoolingRatio(self):
+  def testPoolingRatioHasMoreDimThanInput(self):
     with self.cached_session() as _:
       with self.assertRaisesRegex(
           errors.InvalidArgumentError,
@@ -338,6 +338,23 @@ class FractionalMaxPoolTest(test.TestCase):
             name=None)
         self.evaluate(result)
 
+  def testPoolingRatioValueOutOfRange(self):
+    with self.cached_session() as _:
+      # Whether turn on `TF2_BEHAVIOR` generates different error messages
+      with self.assertRaisesRegex(
+          (errors.InvalidArgumentError, ValueError),
+          r"(pooling_ratio cannot be smaller than 1, got: .*)|(is negative)"):
+        result = nn_ops.gen_nn_ops.fractional_max_pool(
+            value=np.zeros([3, 30, 30, 3]),
+            pooling_ratio=[1, -1, 3, 1],
+            pseudo_random=False,
+            overlapping=False,
+            deterministic=False,
+            seed=0,
+            seed2=0,
+        )
+        self.evaluate(result)
+
 
 class FractionalMaxPoolGradTest(test.TestCase):
   """Tests for FractionalMaxPoolGrad.