CVE-2022-21736: Undefined behavior in Tensorflow
Impact The implementation of SparseTensorSliceDataset has an undefined behavior: under certain condition it can be made to dereference a nullptr value:
python import tensorflow as tf import numpy as np
tf.rawops.SparseTensorSliceDataset( indices=[[]], values=[], denseshape=[1,1])
The 3 input arguments represent a sparse tensor. However, there are some preconditions that these arguments must satisfy but these are not validated in the implementation.
Patches We have patched the issue in GitHub commit 965b97e4a9650495cda5a8c210ef6684b4b9eceb.
The fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.
For more information Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.
Attribution This vulnerability has been reported by Faysal Hossain Shezan from University of Virginia.
Other sources
Tensorflow is an Open Source Machine Learning Framework. The implementation of SparseTensorSliceDataset has an undefined behavior: under certain condition it can be made to dereference a nullptr value. The 3 input arguments to SparseTensorSliceDataset represent a sparse tensor. However, there are some preconditions that these arguments must satisfy but these are not validated in the implementation. The fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.
Affected Software
Remediation
Event History
Frequently Asked Questions
What is the severity of CVE-2022-21736?
CVE-2022-21736 has a medium severity rating due to the potential for undefined behaviors affecting application stability.
How do I fix CVE-2022-21736?
To fix CVE-2022-21736, upgrade to TensorFlow version 2.7.1 or later, or apply the appropriate updates for earlier versions as specified.
What versions of TensorFlow are affected by CVE-2022-21736?
CVE-2022-21736 affects TensorFlow versions up to and including 2.7.0.
Is CVE-2022-21736 related to a specific feature in TensorFlow?
Yes, CVE-2022-21736 is related to the implementation of SparseTensorSliceDataset in TensorFlow.
What should I do if I'm using an affected version of TensorFlow?
If using an affected version of TensorFlow, it is recommended to update to a non-vulnerable version as soon as possible to mitigate the risk.