VYPR
← All advisories
Moderate severityNVD Advisory· Published May 20, 2022· Updated Apr 22, 2025

Missing validation causes denial of service in TensorFlow via `Conv3DBackpropFilterV2`

CVE-2022-29196

Description

TensorFlow is an open source platform for machine learning. Prior to versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4, the implementation of tf.raw_ops.Conv3DBackpropFilterV2 does not fully validate the input arguments. This results in a CHECK-failure which can be used to trigger a denial of service attack. The code does not validate that the filter_sizes argument is a vector. Versions 2.9.0, 2.8.1, 2.7.2, and 2.6.4 contain a patch for this issue.

Affected packages

Versions sourced from the GitHub Security Advisory.

PackageAffected versionsPatched versions
tensorflowPyPI
< 2.6.42.6.4
tensorflowPyPI
>= 2.7.0, < 2.7.22.7.2
tensorflowPyPI
>= 2.8.0, < 2.8.12.8.1
tensorflow-cpuPyPI
< 2.6.42.6.4
tensorflow-cpuPyPI
>= 2.7.0, < 2.7.22.7.2
tensorflow-cpuPyPI
>= 2.8.0, < 2.8.12.8.1
tensorflow-gpuPyPI
< 2.6.42.6.4
tensorflow-gpuPyPI
>= 2.7.0, < 2.7.22.7.2
tensorflow-gpuPyPI
>= 2.8.0, < 2.8.12.8.1

Affected products

1

Patches

1
174c5096f303

Fix failed check in Conv3DBackpropFilterV2.

https://github.com/tensorflow/tensorflowAntonio SanchezApr 29, 2022via ghsa
commit
2 files changed · +30 −0
  • tensorflow/core/kernels/conv_grad_ops_3d.cc+12 −0 modified
    @@ -741,6 +741,10 @@ class Conv3DBackpropFilterOp : public OpKernel {
     TensorShape filter_shape;
     if (takes_shape_) {
       const Tensor& filter_sizes = context->input(1);
+      OP_REQUIRES(context, TensorShapeUtils::IsVector(filter_sizes.shape()),
+                  errors::InvalidArgument(
+                      "filter_sizes shape must be rank 1 but is rank ",
+                      filter_sizes.shape().dims()));
       OP_REQUIRES_OK(context, TensorShapeUtils::MakeShape(
                                   filter_sizes.vec<int32>(), &filter_shape));
     } else {
@@ -875,6 +879,10 @@ class Conv3DCustomBackpropFilterOp : public OpKernel {
     TensorShape filter_shape;
     if (takes_shape_) {
       const Tensor& filter_sizes = context->input(1);
+      OP_REQUIRES(context, TensorShapeUtils::IsVector(filter_sizes.shape()),
+                  errors::InvalidArgument(
+                      "filter_sizes shape must be rank 1 but is rank ",
+                      filter_sizes.shape().dims()));
       OP_REQUIRES_OK(context, TensorShapeUtils::MakeShape(
                                   filter_sizes.vec<int32>(), &filter_shape));
     } else {
@@ -1638,6 +1646,10 @@ class Conv3DBackpropFilterOp<GPUDevice, T> : public OpKernel {
     TensorShape filter_shape;
     if (takes_shape_) {
       const Tensor& filter_sizes = context->input(1);
+      OP_REQUIRES(context, TensorShapeUtils::IsVector(filter_sizes.shape()),
+                  errors::InvalidArgument(
+                      "filter_sizes shape must be rank 1 but is rank ",
+                      filter_sizes.shape().dims()));
       OP_REQUIRES_OK(context, tensor::MakeShape(filter_sizes, &filter_shape));
     } else {
       filter_shape = context->input(1).shape();
  • tensorflow/python/kernel_tests/nn_ops/conv3d_backprop_filter_v2_grad_test.py+18 −0 modified
    @@ -18,6 +18,7 @@
 
 from tensorflow.python.framework import constant_op
 from tensorflow.python.framework import dtypes
+from tensorflow.python.framework import errors
 from tensorflow.python.framework import test_util
 from tensorflow.python.ops import array_ops
 from tensorflow.python.ops import gradient_checker
@@ -58,6 +59,23 @@ def testGradient(self):
           err_tolerance = 1e-3
           self.assertLess(err, err_tolerance)
 
+  def testBadFilterShape(self):
+    strides = [1, 1, 1, 1, 1]
+    padding = "VALID"
+    tin = constant_op.constant(
+        .5053710941, shape=[2, 2, 2, 2, 1], dtype=dtypes.float32)
+    filter_sizes = constant_op.constant(0, shape=[], dtype=dtypes.int32)
+    out_backprop = constant_op.constant(
+        .5053710941, shape=[2, 2, 2, 2, 1], dtype=dtypes.float32)
+
+    with self.assertRaisesRegex((ValueError, errors.InvalidArgumentError),
+                                "must be rank 1"):
+      nn_ops.conv3d_backprop_filter_v2(
+          input=tin,
+          filter_sizes=filter_sizes,
+          out_backprop=out_backprop,
+          strides=strides,
+          padding=padding)
 
 if __name__ == "__main__":
   test.main()

Vulnerability mechanics

Generated by null/stub on May 9, 2026. Inputs: CWE entries + fix-commit diffs from this CVE's patches. Citations validated against bundle.

References

9

News mentions

0

No linked articles in our index yet.