Out-of-Bounds Affecting tensorflow-gpu package, versions [2.7.0,2.7.1)


Severity

Recommended
0.0
high
0
10

CVSS assessment made by Snyk's Security Team. Learn more

Threat Intelligence

EPSS
0.04% (15th percentile)

Do your applications use this vulnerable package?

In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.

Test your applications
  • Snyk IDSNYK-PYTHON-TENSORFLOWGPU-2395359
  • published6 Feb 2022
  • disclosed6 Feb 2022
  • creditUnknown

Introduced: 6 Feb 2022

CVE-2022-23594  (opens in a new tab)
CWE-119  (opens in a new tab)

How to fix?

Upgrade tensorflow-gpu to version 2.7.1 or higher.

Overview

tensorflow-gpu is a machine learning framework.

Affected versions of this package are vulnerable to Out-of-Bounds in the TFG dialect, due to it making several assumptions about the incoming GraphDef before converting it to the MLIR-based dialect. If an attacker changes the SavedModel format on the disk to invalidate these assumptions and the GraphDef is converted to MLIR-based IR, then they can cause a crash in the Python interpreter.

CVSS Scores

version 3.1