PT-2021-23187 · Google · Tensorflow

Mihaimaruseac

·

Published

2021-11-05

·

Updated

2024-03-06

·

CVE-2021-41214

CVSS v3.1

7.8

High

VectorAV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Name of the Vulnerable Software and Affected Versions TensorFlow versions prior to 2.7.0 TensorFlow versions 2.6.0 through 2.6.0 (will be fixed in 2.6.1) TensorFlow versions 2.5.0 through 2.5.1 (will be fixed in 2.5.2) TensorFlow versions 2.4.0 through 2.4.3 (will be fixed in 2.4.4)
Description The shape inference code for tf.ragged.cross has an undefined behavior due to binding a reference to nullptr. This results in a crash when using the tf.ragged.cross function with specific inputs, such as tf.ragged.cross([tf.ragged.constant([['1']]), '2']). The issue is caused by the tf.ragged.cross function and can be exploited in certain scenarios.
Recommendations For TensorFlow versions prior to 2.7.0, update to version 2.7.0 or later. For TensorFlow versions 2.6.0 through 2.6.0, update to version 2.6.1 or later. For TensorFlow versions 2.5.0 through 2.5.1, update to version 2.5.2 or later. For TensorFlow versions 2.4.0 through 2.4.3, update to version 2.4.4 or later. As a temporary workaround, consider avoiding the use of the tf.ragged.cross function until a patch is available.

Exploit

Fix

Access of Uninitialized Pointer

Weakness Enumeration

Related Identifiers

BIT-TENSORFLOW-2021-41214
CVE-2021-41214
GHSA-VWHQ-49R4-GJ9V
OPENSUSE-SU-2024:12116-1
PYSEC-2021-406
PYSEC-2021-623
PYSEC-2021-821

Affected Products

Tensorflow