CVE-2021-29561: Denial of Service in TensorFlow-GPU

Sponsored Habilume Habit tracker with a companion that grows with you Complete habits, earn Lume Points, and watch your Lume grow as you stay consistent. App Store Google Play

Background

TensorFlow is an open-source platform for machine learning that provides a comprehensive ecosystem of tools, libraries, and resources. It is widely used for developing and deploying machine learning models. TensorFlow-GPU is a variant of TensorFlow that leverages GPU acceleration for faster computation. It allows users to train and run machine learning models on GPUs, significantly improving performance. However, TensorFlow-GPU is vulnerable to a denial of service attack due to a flaw in the implementation of tf.raw_ops.LoadAndRemapMatrix.

Vulnerability Detail

The vulnerability arises from an incorrect assumption made in the implementation of tf.raw_ops.LoadAndRemapMatrix. The code assumes that the ckpt_path parameter is always a valid scalar. However, an attacker can exploit this by sending a different tensor as the first argument of LoadAndRemapMatrix. This triggers a rank check, causing a CHECK failure and terminating the process. This vulnerability can be exploited by an attacker to cause a denial of service, disrupting the availability of the TensorFlow-GPU application. The issue has been assigned CVE-2021-29561.

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

Workaround

As a temporary workaround, users can implement input validation to ensure that the ckpt_path parameter passed to LoadAndRemapMatrix is a valid scalar. This can be done by checking the rank of the tensor and raising an error if it is not a scalar. Here is an example of how to perform this validation in Python:

import tensorflow as tf

def load_and_remap_matrix(ckpt_path):
    if tf.rank(ckpt_path) != 0:
        raise ValueError('Invalid ckpt_path')
    tf.raw_ops.LoadAndRemapMatrix(ckpt_path)

Conclusion

The vulnerability in TensorFlow-GPU can be exploited by an attacker to cause a denial of service. It is crucial for users to update their installations to the patched versions as soon as they become available. Regularly checking for updates and applying them promptly is recommended to ensure the security and stability of the application. For further support on vulnerability remediation, please contact DevNack.com.

See every vulnerability recorded for tensorflow-gpu →

FAQ