PT-2022-26142 · Google · Tensorflow
Published
2022-11-18
·
Updated
2024-03-06
·
CVE-2022-41911
CVSS v3.1
4.8
Medium
| Vector | AV:N/AC:H/PR:L/UI:R/S:U/C:N/I:N/A:H |
Name of the Vulnerable Software and Affected Versions
TensorFlow versions prior to 2.11.0
TensorFlow versions 2.10.1 and earlier
TensorFlow versions 2.9.3 and earlier
TensorFlow versions 2.8.4 and earlier
Description
The issue arises when printing a tensor, as the data is retrieved as a
const char* array and then typecast to the element type. However, conversions from char to bool are undefined if the char is not 0 or 1, causing sanitizers and fuzzers to crash. The vulnerability was discovered via internal fuzzing.Recommendations
For TensorFlow versions prior to 2.11.0, update to version 2.11.0 or later.
For TensorFlow versions 2.10.1 and earlier, update to version 2.10.1 or later, or apply the patch from GitHub commit
1be74370327.
For TensorFlow versions 2.9.3 and earlier, update to version 2.9.3 or later, or apply the patch from GitHub commit 1be74370327.
For TensorFlow versions 2.8.4 and earlier, update to version 2.8.4 or later, or apply the patch from GitHub commit 1be74370327.
As a temporary workaround, consider avoiding the printing of tensors that may contain boolean values until the issue is resolved.Exploit
Fix
Incorrect Type Conversion or Cast
Found an issue in the description? Have something to add? Feel free to write us 👾
Weakness Enumeration
Related Identifiers
Affected Products
Tensorflow