CVE-2022-21734: TensorFlow CHECK Failures Due to Non-Scalar Key Tensors
Summary
CVE-2022-21734: CHECK-failures in Tensorflow. Affects tensorflow across 3 version ranges.
< 2.5.3
>= 2.6.0, < 2.6.3
>= 2.7.0, < 2.7.1
Fixed in 2.5.3, 2.6.3 or 2.7.1, depending on the release line you run. How to fix ↓
Also tracked for
- tensorflow · Bitnami
- tensorflow-gpu · PyPI
- tensorflow-cpu · PyPI
The same CVE-2022-21734 is tracked separately for each ecosystem and vendor. Its fix information is the same; the package and affected versions differ.
Check if this vulnerability affects your application
Upload a manifest or SBOM and we'll show whether you're affected, plus every other risky package in the same scan.
No signup required · Results in about 60 seconds
Background
TensorFlow is an open-source machine learning framework that provides a comprehensive ecosystem for building and deploying machine learning models. It is widely used for its flexibility and scalability, allowing developers to create complex neural networks and perform various computations efficiently. However, like any software, TensorFlow is susceptible to vulnerabilities that can affect its functionality and security. This specific vulnerability arises from the implementation of the MapStage operation, which can lead to unexpected behavior if the key tensor is not a scalar.
Vulnerability Detail
The vulnerability in TensorFlow is caused by a CHECK failure in the MapStage operation when the key tensor is not a scalar. This can lead to application crashes or undefined behavior, potentially compromising the integrity of the application using TensorFlow. The issue originates from the way the MapStage operation validates its input parameters, specifically the key tensor. If a non-scalar tensor is provided, it triggers a CHECK failure, which can disrupt the execution of the program. This vulnerability affects multiple versions of TensorFlow, making it critical for users to address it promptly.
How To Fix
Latest published version: 2.21.0 as of 22 Oct 2026
pip install --upgrade "tensorflow==2.21.0"Upgrading to it may include breaking changes; the versions below are the minimum fixed release for this vulnerability.
If you can't move to the latest
The minimum fixed version depends on the release line you run.
-
< 2.5.32.5.3pip install --upgrade "tensorflow==2.5.3" -
>= 2.6.0, < 2.6.32.6.3pip install --upgrade "tensorflow==2.6.3" -
>= 2.7.0, < 2.7.12.7.1pip install --upgrade "tensorflow==2.7.1"
Tip — Not sure which release line you're on? Vulert can find the fix for the version you are using. Scan free →
Workaround
As a workaround, ensure that the key tensor passed to the MapStage operation is always a scalar. For example, instead of passing a non-scalar tensor, use the following code:
import tensorflow as tf import numpy as np
tf.raw_ops.MapStage( key=tf.constant(value=4, dtype=tf.int64), # Ensure key is a scalar indices=np.array([[6]]), values=np.array([-60]), dtypes=[tf.int64], capacity=0, memory_limit=0, container='', shared_name='', name=None
)
Conclusion
The CHECK failure vulnerability in TensorFlow's MapStage operation poses a risk to applications using non-scalar key tensors. It is crucial for users to update to the patched versions of TensorFlow to mitigate this risk. By following the provided steps and workarounds, developers can ensure their applications remain secure.
See every vulnerability recorded for tensorflow →
FAQ
The vulnerability can lead to application crashes or undefined behavior when a non-scalar key tensor is used in the MapStage operation.
The vulnerability affects TensorFlow versions prior to 2.8.0, including 2.5.3, 2.6.0, and 2.7.0.
You can use Vulert to monitor and alert for open-source vulnerabilities in your software. Visit Vulert Scanner to check if your application is affected without any signup.
As a temporary measure, ensure that the key tensor used in the MapStage operation is a scalar to avoid triggering the CHECK failure.
This vulnerability was reported by Faysal Hossain Shezan from the University of Virginia.
References
Related vulnerabilities
Recent vulnerabilities recorded for tensorflow.
