CVE-2022-21728: Out of Bounds Read in Tensorflow

Background

Tensorflow is an open-source machine learning framework developed by Google. It provides a comprehensive ecosystem of tools, libraries, and resources for building and deploying machine learning models. Tensorflow is widely used in various domains, including research, industry, and academia. However, a vulnerability has been discovered in Tensorflow that could potentially lead to an out of bounds read.

Vulnerability Detail

The vulnerability in Tensorflow is related to the implementation of shape inference for the ReverseSequence operation. The code does not properly validate the value of the batch_dim parameter, which can result in a heap out of bounds read. An attacker could exploit this vulnerability to read sensitive information from the heap, potentially leading to information disclosure or other security issues. The vulnerability has a CVSS score of 8.1, indicating a high severity level.

pip install --upgrade "tensorflow==2.21.0"

Workaround

As a temporary workaround, you can avoid using the ReverseSequence operation with negative values for the seq_dim and batch_dim parameters. Make sure to validate and sanitize user input to prevent any potential exploitation of this vulnerability. Here is an example of how to modify the code to avoid the vulnerability:

import tensorflow as tf

@tf.function
def test():
  y = tf.raw_ops.ReverseSequence(
    input=['aaa', 'bbb'],
    seq_lengths=[1, 1, 1],
    seq_dim=0,  # Modify the value of seq_dim
    batch_dim=0  # Modify the value of batch_dim
  )
  return y

test()

Conclusion

In conclusion, the vulnerability in Tensorflow could lead to an out of bounds read, potentially resulting in information disclosure or other security issues. It is crucial to update to the latest patched version of Tensorflow to mitigate the risk. If you are unable to update immediately, consider implementing the provided workaround and ensure proper input validation and sanitization. For further support on vulnerability remediation, please contact DevNack.com.

See every vulnerability recorded for tensorflow →

FAQ