CVE-2021-29569: TensorFlow GPU - Out-of-Bounds Read in MaxPoolGradWithArgmax

Background

TensorFlow is an end-to-end open source platform for machine learning. It provides a comprehensive ecosystem of tools, libraries, and resources for developing and deploying machine learning models. TensorFlow GPU is a variant of TensorFlow that leverages GPU acceleration for faster computation. It is widely used in various industries and research domains for deep learning tasks.

Vulnerability Detail

The vulnerability (CVE-2021-29569) exists in the implementation of tf.raw_ops.MaxPoolGradWithArgmax in TensorFlow GPU. By supplying specially crafted inputs, an attacker can trigger an out-of-bounds read, leading to potential information disclosure or denial of service. The issue arises from the assumption made in the implementation that the input_min and input_max tensors have at least one element. If these tensors are empty, accessing the 0th element results in a read outside the bounds of the allocated data. The vulnerability affects versions of TensorFlow GPU from 0 to 2.1.4, 2.2.0 to 2.2.3, 2.3.0 to 2.3.3, and 2.4.0 to 2.4.2.

pip install --upgrade "tensorflow-gpu==2.12.0"

Workaround

As a workaround, users can avoid the vulnerability by ensuring that the input_min and input_max tensors passed to tf.raw_ops.MaxPoolGradWithArgmax have at least one element. This can be achieved by validating the input tensors before invoking the vulnerable function. Here's an example in Python:

import tensorflow as tf

# Validate input tensors
if len(input_min) == 0 or len(input_max) == 0:
    # Handle empty tensors
    print('Input tensors must have at least one element.')
else:
    # Invoke the vulnerable function
    result = tf.raw_ops.MaxPoolGradWithArgmax(input_min, input_max)

Conclusion

In conclusion, the vulnerability (CVE-2021-29569) in TensorFlow GPU's tf.raw_ops.MaxPoolGradWithArgmax function can result in an out-of-bounds read if attacker-controlled inputs are supplied. The TensorFlow development team has addressed the issue in the latest versions, and users are advised to upgrade to mitigate the vulnerability. It is crucial to keep software dependencies up to date and follow best practices for secure development. For further support on vulnerability remediation, you can reach out to DevNack.com.

See every vulnerability recorded for tensorflow-gpu →

FAQ